`, the title lines, `W`/`H`, and fill `nodes[]`, `edges[]`, and optionally `BANDS[]`, `COLUMNS{}` / `HCOLUMNS[]`, `CANVASES[]`, `STATUS_LEGEND{}`. Minimal shape (full fields: `schema.md § Node` / `§ Edge`): ```js // node — required: id, label, cat, type, desc, x, y { id:'FEED', label:'Feeding', cat:'CORE LOOP', type:'core', role:'action', desc:'Tap to feed.\nRestores hunger.', x:220, y:120 } // edge — required: f, t, fromPt, toPt { f:'FEED', t:'FOOD', fromPt:'bot', toPt:'top', route:'vhv', style:'solid' } ``` Declare node ownership on **both sides** — a node's `band`/`free` (plus `column`/`colFree` in H) and the band/column `members[]` must agree (gated A27/A28). The engine does the rest: bands auto-fit, lanes auto-grow, loose entry nodes wrap in a LIFECYCLE band. (`schema.md § Node ownership` · `§ Column ownership` · `§ Auto LIFECYCLE band` · `§ Canvas filter`.) ### 5. Features & styling — all optional, all in the references Add only what the flow needs; every field is defined in `schema.md`, every heuristic in `layout-tips.md`: - **Node types** — built-ins + the custom-type recipe: `§ Adding a custom type`. - **Edge routing** — `vhv`/`hvh`/`3leg` per edge; pick `fromPt`/`toPt` so lines exit/enter clean (arrowhead points at `toPt`): `§ Routers` + `layout-tips.md §6, §9`. - **Edge styling** — solid/dashed/bold + `crit`: `§ Edge styling decision matrix`. - **Critical paths** — `crit:true` + `lbl:'🔑N +verb'`, 3–7 per 30 edges: `layout-tips.md §8`. - **Decisions** — `kind:'diamond'` + `branch:'Y'/'N'`: `§ Decision diamond`. - **Status dots + author flags** — `status` (+ `STATUS_LEGEND`) and `flag` (a lightweight per-node "look here" note): `§ Status dot` · `§ Audit marker`. For a real **defect review** — bugs, logic holes, geometry defects like a wire through a node — use the **Audit** extension, not `flag` icons (§ Audit). ### 6. Render — three modes (fall back in order; don't skip ahead) - **Mode 1 — inline interactive widget.** Render `template-v2.html` via the host's HTML tool (e.g. `mcp__visualize__show_widget`). Pan/zoom/hover native — best for iteration. - **Mode 2 — inline static SVG (your own output, not a tool call).** If Mode 1 is unavailable, write an `<svg>…</svg>` block directly in your reply, using arcgram conventions (bands as rects, nodes border+dark bg, `vhv`/`hvh`/`3leg` edges, pills, crit red). Don't switch to a different diagram tool — the chat-native SVG routes *around* the failed tool. - **Mode 3 — standalone HTML export.** Write `template-v2.html` with your data to a file the user opens. Always available with filesystem access; use for handoff. ### 7. Self-check — run the validators (headless Node, under `skills/<tool>/`) The diagram is a *spec* — check it, don't eyeball the render. **Pre-draw** (as soon as `nodes`/`edges` exist) — Checkpoint catches dangling edges, orphans, a diamond missing a branch, duplicate ids: ```bash node skills/checkpoint/checkpoint.mjs your.html ``` **Post-draw** — Validate runs Checkpoint always, and Reconcile too when you pass matching code (a no-code run reports INCOMPLETE, never a silent pass): ```bash node skills/validate/validate.mjs your.html [code-file ...] ``` The engine also self-checks on render (stamps `self-check ran … | N findings` into the subtitle); the CLI is the enforce-by-construction version. ### 8. Save + syntax-check (Mode 3) Extract the inline JS and syntax-check before opening — catches typos that silently blank the canvas: ```bash awk '/<script>/{flag=1;next}/<\/script>/{flag=0}flag' your.html > /tmp/x.js && node --check /tmp/x.js ``` Open in browser. Confirm. --- ## Audit — defect review (opt-in; use the extension, not `flag` icons) Audit is a defect-review overlay for a *drawn* flow: a pulsing ring + a short reason tag on each marked node or edge, plus an "Audit" findings list. It **self-hides when there are no marks**, so a clean flow looks identical with or without it. **Ask first, mark only when asked.** Offer the user an Audit pass and settle it up front — do not auto-annotate defects. **Use the extension — not the `flag` field.** The Audit UI is `extensions/arcgram-bugmarks.js`. The per-node `flag` field draws a small icon but is *not* the Audit overlay — it must never stand in for a defect review. Load the extension *after* the engine (inline the `<script>` for a self-contained file), then call `ArcgramBugs.set([...])`: ```html <script src="extensions/arcgram-bugmarks.js"></script> <script> ArcgramBugs.set([ { id:'BUG-260911-ag-01', level:'high', type:'logic', anchor:{ node:'NODE_ID' }, reason:'no failure branch', desc:'a declined payment dead-ends here — the path assumes success', evidence:'scanner: 1 outgoing edge (success), 0 failure branch' }, { id:'BUG-260911-ag-02', level:'medium', type:'logic', anchor:{ edge:{ f:'A', t:'B' } }, reason:'unstated precondition', desc:'…', evidence:'doc: the spec requires a confirmed session here' } ]); </script> ``` `level` is **severity — how bad it is**, not where it came from: **`high`** = the flow is wrong and someone will hit it, **`medium`** = a real concern worth a decision, **`low`** = an advisory. (The checkpoint skill's `severity` field is a *different* measurement that happens to share the word — never map one onto the other.) Where a mark came from goes in `evidence`, prefixed `scanner:` / `doc:` / `derived:`. Mark **reasoning**, not geometry: use `type:'logic'`. Geometry is gated, not annotated — the gate bundle checks it and the engine fixes it. Anchor each mark to a `node:'ID'` or an `edge:{ f, t }`, give it an id shaped `BUG-<yymmdd>-<agent>-<seq>`, and ground it with `evidence`. Full API, the closure fields, and the id rules: `extensions/README.md`. **Closed marks on the flow — an authored per-map default.** Resolved rings are hidden by default (the header then says "N closed hidden"); a map whose closure evidence is part of the story declares `` before the extension's script region and its closed rings show from the first load. Settle it with the user like the Audit pass itself. Stamp semantics + the unknown-stamp allow-list: `extensions/README.md § Per-map stamps`. --- ## Hard rules Non-negotiables. Adjust everything else freely; these hold on every diagram. - **Write permission (two halves).** The agent edits **only the DATA SECTION** — everything below `END OF DATA SECTION` is the engine, off-limits. AND the agent **proposes; the human holds final write-authority** — draft and adjust the diagram data freely, but do not commit it as final shared ground truth without the human's confirmation. - **Author from `schema.md`, never from memory** of what generic diagram tools look like. Read this body + `schema.md` before generating; name fields, don't invent them. - **Single source.** `schema.md` owns every field; `README.md` owns the positioning. Point to them — don't restate a field definition or re-argue the job inside this file. - **Keep attribution.** The "Made with Arcgram" mark + `NOTICE` stay intact when you redistribute (Apache §4(d)); an unbranded / no-attribution build needs the commercial license (`WATERMARK-AND-COMMERCIAL-TERMS.md`). --- ## License **Apache-2.0.** Attribution rules → `§ Hard rules` (the canonical copy); full terms, trademark, and the unbranded commercial build → `LICENSE` · `NOTICE` · `WATERMARK-AND-COMMERCIAL-TERMS.md`. 22:o8,27:Td7e,22:od7e,--- name: checkpoint description: >- Validate an Arcgram diagram as a logic spec: is the flow well-formed and coherent? Reads node/edge structure (id/f/t/kind/role/type), never geometry. Typed findings - dangling edge, orphan node, missing branch, disconnected island, duplicate id, missing schema, plus optional coherence (every action traces to a goal). Marks, never blocks. Run after authoring or editing a diagram. Agent-neutral. Sibling to Reconcile (Checkpoint = structure; Reconcile = values). --- # Checkpoint - Arcgram logic self-check validator One input: a **diagram** (treated as a spec). One question: **is the flow logic well-formed and coherent?** Returns typed findings; marks, never blocks. Sibling to Reconcile: Checkpoint checks **structure** from the diagram alone; Reconcile checks **values** across diagram + code. Run both -> the diagram is a faithful, current spec. It checks the graph's logic structure - **not** drawing quality (the engine auto-handles overlaps/spacing/routing) and **not** semantic correctness (whether the logic is *right* - a human call). **Red line:** reads only the semantic field values `id`, `f`, `t`, `kind`, `role`, `type` (plus optional `status`, `flag`). Never a coordinate, color, size, or spacing. The judge: am I reading a semantic string, or a rendered visual quantity? ## Findings **Tier A - structural integrity (always on).** Domain-agnostic graph theory. | finding | means | |---|---| | `dangling-edge` | an edge points at an id that is not a node | | `orphan-node` | a node has no edge (chain broken) - except a declared `type:'open'` | | `missing-branch` | a decision (`kind:'diamond'`) lacks its Y or N branch | | `disconnected-island` | a cluster with no edge to the main flow | | `duplicate-id` | a node id used more than once | | `missing-field` | a node has no id, or an edge no `f`/`t` | | `missing-meta-version` | `ARCGRAM_META.version` absent | | `empty-diagram` | no nodes - reported uncovered, never falsely "clean" | **Tier B - justification coherence (auto-on for a thinking flow).** Does it read as reasoning: every `action` traces back to a `goal`, every `goal` reaches an `action`, every `decision` sits on a justification path. Auto-on when the diagram has a decision diamond (`kind:'diamond'`, = the engine's `_isThinkflow`); no-diamond diagrams skip it. Override `--tier-b` / `--no-tier-b`. Roles: `goal`/`action`/`decision` (coherence-bearing) + `constraint`/`evidence`/`state`/`output` (recognized). No resolvable role -> `[?]` uncovered, never guessed. `cat` is not read. Each finding is `{ id, type, severity, note }`; severity = `defect` or `uncovered`. ## Output (same contract as Reconcile) ``` { findings:[{id,type,severity,note}], summary, attestation, clean } self-check ran HH:MM:SS | N findings | M unchecked ``` **clean only when N=0 AND M=0** - "checked & clean" never looks like "never checked". Human reads the marker; agent reads the `type`. ## Use ```sh node checkpoint.mjs diagram.html [--tier-b | --no-tier-b] [--json] cat diagram.json | node checkpoint.mjs ``` ```js import { checkpoint } from './checkpoint.mjs'; checkpoint({ nodes, edges, meta }); ``` Exit 1 on any defect (advisory - never blocks). Verify: `node checkpoint-selftest.mjs` (19 checks). Render-time: the engine exposes the same check via `window.__arcgramInspect()` (the `checkpoint` field of its result) + the on-canvas `[?]` badge and attestation stamp. 22:o8,28:Td92,22:od92,--- name: reconcile description: >- Check that an Arcgram diagram's @spec values still match the code that implements it. Reads each node's desc "@spec key=value" line, finds the same-named code constant (hold_ms <-> HOLD_MS / holdMs), diffs the literals. Typed findings: drift, missing constant, non-literal, ambiguous. Reads only the @spec line + named constants, never geometry; marks, never blocks. Run after editing a diagram or its code. Agent-neutral. Sibling to Checkpoint (Checkpoint = structure; Reconcile = values). --- # Reconcile - code<->diagram @spec bijection validator Two inputs: a **diagram** + the **code** it describes. One question: **do the diagram's `@spec` values still equal the same-named code constants?** Returns typed findings; marks, never blocks. Sibling to Checkpoint: Checkpoint checks **structure** from the diagram alone; Reconcile checks **values** across diagram + code. Run both -> the diagram is a faithful, current spec. ## Input: the `@spec` line A node pins values on one line at the top of `desc`: ``` @spec hold_ms=800 fade_ms=200 <- Reconcile reads ONLY this line ``` - Prefix `@spec ` (ASCII). `key=value`, **key = the code constant name**. - Match is case/separator-insensitive: `hold_ms` = `HOLD_MS` = `holdMs` = `hold-ms`. - Value: a number (`800`, `0x320`) or a quoted string (`"left"`). - Free prose and a `~ tune: <range>` line are for humans - never read. Reconcile reads **only** the `@spec` line + **named** code constants. No geometry, no prose, no program behavior. ## Findings | finding | severity | means | |---|---|---| | `spec-drift` | defect | @spec value != the code constant | | `spec-no-constant` | defect | no same-named constant in the code | | `malformed-spec` | defect | @spec line is not `key=value` | | `value-uncovered` | uncovered | constant is computed, not a literal - no verdict | | `ambiguous-constant` | uncovered | two constants normalize to one name | | `no-spec-lines` | uncovered | nothing pinned - never a false "clean" | **Anchored on the `@spec` line** (not the code): drift is caught both ways, but a code constant on no `@spec` line is normal, not a finding. The extractor is a tolerant, dependency-free scan of `const/let/var =`, `NAME =`, `NAME:`, `#define`, `:=`, `[modifiers] const|readonly TYPE NAME =`, and C#/Unity **bare field initializers** `[attr] public|private|static TYPE NAME = value` (ScriptableObject tuning fields, no const/readonly) (JS/TS/Python/C/Go/C#/Java); numbers normalize (`800` == `800.0` == `0x320`, and a C#/Java numeric suffix is dropped so `0.2f` == `0.2`, `800L` == `800`). A Unity bare field is matched only with a field-only marker (an access modifier, `[Attribute]`, or `static`), so a method-body local (`int x = 5;`) is never captured. ## Output (same contract as Checkpoint) ``` { findings:[{id,type,severity,note}], summary, attestation, clean } self-check ran HH:MM:SS | N findings | M unchecked ``` **clean only when N=0 AND M=0** - "checked & clean" never looks like "never checked". Human reads the marker; agent reads the `type`. ## Use ```sh node reconcile.mjs diagram.html code.js [more.js ...] [--json] ``` ```js import { reconcile } from './reconcile.mjs'; reconcile({ nodes }, codeText); ``` Exit 1 on any defect (advisory - never blocks). Verify: `node reconcile-selftest.mjs` (48 checks). Example: `../../examples/example-workflow.html` vs `example-workflow.config.cs` (clean, reconciles 0 findings). 22:o9,29:T1073,22:o1073,--- name: validate description: >- Run both Arcgram self-checks in one call. A thin wrapper over Checkpoint (structure) and Reconcile (values): validate(diagram,{code}) runs Checkpoint always and Reconcile when code is supplied, then merges into one source-tagged result in the same {findings,summary,attestation,clean} contract. clean = checkpoint.clean && reconcile.clean. Adds no new check; marks, never blocks. Run after authoring or editing a diagram and its code. Agent-neutral. The single door over its two siblings. --- # Validate - the combined Arcgram self-check entry One call. Two questions answered between the siblings: **is the flow well-formed and coherent, AND do its `@spec` values still match the code?** Validate is a **thin wrapper**, not a new check. It imports [Checkpoint](../checkpoint/SKILL.md) and [Reconcile](../reconcile/SKILL.md) **untouched** and composes their pure functions: - **Checkpoint** (one input: the diagram) checks **structure** - well-formed, coherent. - **Reconcile** (two inputs: diagram + code) checks **values** - `@spec` literals == code constants. Run both -> the diagram is a faithful, current spec. Every finding still comes from one sibling; Validate only **merges, source-tags, and unions the attestation**. ## Code is optional | input | what runs | clean possible? | |---|---|---| | `validate(diagram, { code })` | Checkpoint **and** Reconcile | yes | | `validate(diagram)` (no code) | Checkpoint only | no - values unchecked | Without code there is nothing to reconcile, so Reconcile is **not run** (running it would just flag every `@spec` pin as missing). Instead the value half is reported as one `uncovered` `value-check-not-run` finding, so a structure-only run is never falsely "clean" on values - the same anti-silent-pass rule both siblings enforce. **Locked invariant - `uncovered` is a third state, never a pass.** A diagram-only run (no code) can **never** produce an overall green clean, even when the structure is perfect. The attestation explicitly appends `| code consistency: NOT CHECKED (no code supplied)`, `summary.codeConsistency` reads `"NOT_CHECKED"`, and the CLI verdict is `INCOMPLETE`, never the word CLEAN. This guards the exact silent-pass the family exists to prevent: claiming "checked & clean" while the code half was never checked. ## Output (same contract as both siblings, plus `source`) ``` { findings:[{id,type,severity,note,source}], summary, attestation, clean } self-check ran HH:MM:SS | N findings | M unchecked ``` - **`source`** is `"checkpoint"` | `"reconcile"` | `"validate"` - so you can route a finding back to the check that raised it. Everything else (`id`, `type`, `severity`, `note`) is verbatim from the sibling. - **`attestation`** is the **union**: N and M are the combined defect / uncovered totals. When the value half did not run it appends an explicit `| code consistency: NOT CHECKED (no code supplied)`. - **`clean === checkpoint.clean && reconcile.clean`** (identically `N=0 && M=0`); `uncovered` is a third state and is never folded into `clean`, so a skipped value-check is not clean. - **`summary`** nests each sibling's own summary under `.checkpoint` / `.reconcile`, plus `ran` and `codeConsistency` (`"checked"` | `"NOT_CHECKED"`). - The full sub-results are returned verbatim on `.checkpoint` / `.reconcile` for a caller that wants the per-check breakdown. Human reads the marker; agent reads the `type` (and `source`). ## Use ```sh node validate.mjs <diagram.html | diagram.json> [code-file ...] [--tier-b | --no-tier-b] [--json] ``` ```js import { validate } from './validate.mjs'; validate({ nodes, edges, meta }, { code }); // code: string | string[]; omit -> structure only ``` `--tier-b` / `--no-tier-b` forward to Checkpoint's justification tier (else auto-detected by a decision diamond). Exit 1 on any defect (advisory - never blocks; an uncovered value-check is not a defect). Verify: `node validate-selftest.mjs` (36 checks - source tags, union attestation, clean rule, the code-omitted path, and byte parity with running the two siblings separately). Example: `examples/example-workflow.html` vs `example-workflow.config.cs` (clean). 22:oc51,11:[["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@type\":\"SoftwareApplication\",\"name\":\"Arcgram\",\"description\":\"Surface an AI agent's thinking and make it auditable. Instead of a wall of prose, your agent commits its reasoning to typed nodes and explicit edges — real data you inspect and correct node by node, converging on ground truth that carries across turns, sessions, and agents. Use it as an agent harness or operating frame: propose → draw → self-audit → fix, human in the loop. Delivered as a single st\",\"applicationCategory\":\"AI assistant add-ons\",\"operatingSystem\":\"Claude\",\"url\":\"https://agentready.co/connectors/arcgram\",\"sameAs\":[],\"publisher\":{\"@type\":\"Organization\",\"name\":\"jovesun-lab\",\"url\":\"https://github.com\"}}"}}],["$","div",null,{"className":"wrap","children":[["$","div",null,{"className":"pp-head","children":[["$","$L23",null,{"name":"Arcgram","url":null,"size":56}],["$","div",null,{"children":[["$","div",null,{"className":"crumbs","children":[["$","$L5",null,{"href":"/","children":"Connectors"}],[" / ",["$","$L5",null,{"href":"/categories/ai-tools","children":"AI assistant add-ons"}]]]}],["$","h1",null,{"children":["Arcgram",null]}]]}]]}],["$","div",null,{"className":"pp","children":[["$","div",null,{"className":"ca","children":[["$","p",null,{"className":"tag","children":"Surface an AI agent's thinking and make it auditable."}],["$","$L24",null,{"text":"Instead of a wall of prose, your agent commits its reasoning to typed nodes and explicit edges — real data you inspect and correct node by node, converging on ground truth that carries across turns, sessions, and agents. Use it as an agent harness or operating frame: propose → draw → self-audit → fix, human in the loop. Delivered as a single standalone interactive HTML file — no build step, no npm, no CDN, no runtime; open in any browser."}],["$","$L25",null,{"examples":[],"tools":[],"prefix":"","server":null,"instructions":[],"skills":[{"name":"arcgram","label":"arcgram","gist":"Generate an interactive node-and-edge diagram as a single standalone HTML file (Arcgram v2).","source":"https://github.com/jovesun-lab/arcgram/blob/fe4fc0eb7d0c6039e2051d02929e28ae5fd12252/SKILL.md","raw":"$26"},{"name":"checkpoint","label":"checkpoint","gist":"Validate an Arcgram diagram as a logic spec: is the flow well-formed and coherent?","source":"https://github.com/jovesun-lab/arcgram/blob/fe4fc0eb7d0c6039e2051d02929e28ae5fd12252/skills/checkpoint/SKILL.md","raw":"$27"},{"name":"reconcile","label":"reconcile","gist":"Check that an Arcgram diagram's @spec values still match the code that implements it.","source":"https://github.com/jovesun-lab/arcgram/blob/fe4fc0eb7d0c6039e2051d02929e28ae5fd12252/skills/reconcile/SKILL.md","raw":"$28"},{"name":"validate","label":"validate","gist":"Run both Arcgram self-checks in one call.","source":"https://github.com/jovesun-lab/arcgram/blob/fe4fc0eb7d0c6039e2051d02929e28ae5fd12252/skills/validate/SKILL.md","raw":"$29"}],"skillNote":"4 in the Claude listing"}],null]}],"$L2a"]}]]}]] 22:oa3d,1b:[["$","title","0",{"children":"Arcgram by jovesun-lab: AI connector · Agent Ready"}],["$","meta","1",{"name":"description","content":"Surface an AI agent's thinking and make it auditable. Instead of a wall of prose, your agent commits its reasoning to typed nodes and explicit edges — real…"}],["$","link","2",{"rel":"manifest","href":"/manifest.webmanifest","crossOrigin":"$undefined"}],["$","link","3",{"rel":"canonical","href":"https://agentready.co/connectors/arcgram"}],["$","meta","4",{"property":"og:title","content":"Arcgram by jovesun-lab: AI connector"}],["$","meta","5",{"property":"og:description","content":"Surface an AI agent's thinking and make it auditable. Instead of a wall of prose, your agent commits its reasoning to typed nodes and explicit edges — real…"}],["$","meta","6",{"property":"og:url","content":"https://agentready.co/connectors/arcgram"}],["$","meta","7",{"property":"og:site_name","content":"Agent Ready"}],["$","meta","8",{"property":"og:image","content":"https://agentready.co/connectors/arcgram/opengraph-image?3b61e23efdbb5225"}],["$","meta","9",{"property":"og:image:type","content":"image/png"}],["$","meta","10",{"property":"og:image:width","content":"1200"}],["$","meta","11",{"property":"og:image:height","content":"630"}],["$","meta","12",{"property":"og:image:alt","content":"AI connector on Agent Ready"}],["$","meta","13",{"property":"og:type","content":"website"}],["$","meta","14",{"name":"twitter:card","content":"summary_large_image"}],["$","meta","15",{"name":"twitter:title","content":"Arcgram by jovesun-lab: AI connector"}],["$","meta","16",{"name":"twitter:description","content":"Surface an AI agent's thinking and make it auditable. Instead of a wall of prose, your agent commits its reasoning to typed nodes and explicit edges — real…"}],["$","meta","17",{"name":"twitter:image","content":"https://agentready.co/connectors/arcgram/opengraph-image?3b61e23efdbb5225"}],["$","meta","18",{"name":"twitter:image:alt","content":"AI connector on Agent Ready"}],["$","meta","19",{"name":"twitter:image:type","content":"image/png"}],["$","meta","20",{"name":"twitter:image:width","content":"1200"}],["$","meta","21",{"name":"twitter:image:height","content":"630"}],["$","link","22",{"rel":"icon","href":"/favicon.ico?favicon.3iymezf4ycw3e.ico","sizes":"64x64","type":"image/x-icon"}],["$","link","23",{"rel":"icon","href":"/icon.svg?icon.1m07wgr9jqhm5.svg","sizes":"any","type":"image/svg+xml"}],["$","link","24",{"rel":"apple-touch-icon","href":"/apple-icon.png?apple-icon.3q8qrgufoflqu.png","sizes":"180x180","type":"image/png"}],["$","$L2b","25",{}]] 22:o10,1c:null 22:null 22:o22,:HL["/stores/claude.png","image"] 22:o128b,2a:["$","aside",null,{"className":"side","children":[["$","div",null,{"className":"card","children":[["$","h2",null,{"children":"Stores"}],["$","div",null,{"className":"st","children":[["$","div","claude",{"style":{"display":"contents"},"children":[["$","span",null,{"className":"s","children":[["$","img",null,{"className":"sm","src":"/stores/claude.png","alt":"","width":16,"height":16}],["$","b",null,{"children":"Claude"}],["$","a",null,{"className":"src","href":"https://github.com/jovesun-lab/arcgram","target":"_blank","rel":"noopener nofollow","children":"Source ↗"}]]}],["$","span",null,{"className":"rk2","children":[["$","small",null,{"children":"Listed, not ranked"}],null]}]]}]]}]]}],["$","div",null,{"className":"card","children":[["$","h2",null,{"children":"Details"}],["$","dl",null,{"className":"kv","children":[[["$","dt",null,{"children":"Company"}],["$","dd",null,{"children":"jovesun-lab"}]],[["$","dt",null,{"children":"Publisher"}],["$","dd",null,{"children":"Community"}]],["$","dt",null,{"children":"Category"}],["$","dd",null,{"children":["$","$L5",null,{"href":"/categories/ai-tools","children":"AI assistant add-ons"}]}],[["$","dt",null,{"children":"Audience"}],["$","dd",null,{"children":"For developers"}]],null,["$","dt",null,{"children":"Sign-in"}],["$","dd",null,{"children":"None needed"}],["$","dt",null,{"children":"Interactive UI"}],["$","dd",null,{"children":"No"}],[["$","dt",null,{"children":"Checked"}],["$","dd",null,{"children":"Oct 27, 2026"}]],null]}]]}],["$","div",null,{"className":"card","children":[["$","h2",null,{"children":["More in ","AI assistant add-ons"]}],["$","ul",null,{"className":"sim","children":[["$","li","14965",{"children":["$","$L5",null,{"href":"/connectors/rhythm-ai-personality-tuner","prefetch":false,"children":[["$","$L23",null,{"name":"Rhythm AI Personality Tuner","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/02/02846ae172d38bae727fc11c13349772f8d2840b740c604ba8c165bdd8d844a9-64.webp","size":24}],["$","b",null,{"children":"Rhythm AI Personality Tuner"}],["$","small",null,{"children":"#146"}]]}]}],["$","li","4302",{"children":["$","$L5",null,{"href":"/connectors/playmcp","prefetch":false,"children":[["$","$L23",null,{"name":"PlayMCP","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/7e/7e9da62ddba915cd96ea88590aa0dd924d416591cf546e27a8d6cb3fe821608b-64.webp","size":24}],["$","b",null,{"children":"PlayMCP"}],["$","small",null,{"children":"#152"}]]}]}],["$","li","15060",{"children":["$","$L5",null,{"href":"/connectors/prompt-perfect","prefetch":false,"children":[["$","$L23",null,{"name":"Prompt Perfect","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/25/25c8b7a6dc543f28b9684227604bf15cec842945f4ac99b7c7fc130d321a8a39-64.webp","size":24}],["$","b",null,{"children":"Prompt Perfect"}],["$","small",null,{"children":"#333"}]]}]}],["$","li","15246",{"children":["$","$L5",null,{"href":"/connectors/basic-memory-cloud","prefetch":false,"children":[["$","$L23",null,{"name":"Basic Memory Cloud","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/48/48765542121cf50a22435c94e535acdf031b1b4208770d51e61a2e1e619acc59-64.webp","size":24}],["$","b",null,{"children":"Basic Memory Cloud"}],["$","small",null,{"children":"#1,199"}]]}]}],["$","li","15582",{"children":["$","$L5",null,{"href":"/connectors/tasknotify","prefetch":false,"children":[["$","$L23",null,{"name":"TaskNotify","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/cc/ccef65bdab248018fd966ac128ac1efec782e1b012f181200f01cb5d4b04753a-64.webp","size":24}],["$","b",null,{"children":"TaskNotify"}],["$","small",null,{"children":"#476"}]]}]}],["$","li","15388",{"children":["$","$L5",null,{"href":"/connectors/app-atlas","prefetch":false,"children":[["$","$L23",null,{"name":"App Atlas","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/8f/8fce6de54fe02b96dd3c7c81562f4faa5a36c4b92a47080da489b59a10709b46-64.webp","size":24}],["$","b",null,{"children":"App Atlas"}],["$","small",null,{"children":"#560"}]]}]}],["$","li","15337",{"children":["$","$L5",null,{"href":"/connectors/caveman-mode","prefetch":false,"children":[["$","$L23",null,{"name":"Caveman Mode","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/27/27efe08ae01ebda82f25893226e33a2c9c3d49b6c6ec787b9a0bff471a580795-64.webp","size":24}],["$","b",null,{"children":"Caveman Mode"}],["$","small",null,{"children":"#604"}]]}]}],["$","li","15406",{"children":["$","$L5",null,{"href":"/connectors/engram-2","prefetch":false,"children":["$L2c","$L2d","$L2e"]}]}]]}]]}]]}] 22:o129,2c:["$","$L23",null,{"name":"Engram","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/37/37541cf9765efdad0155df1c3af9d933e5cd8be560b306e7281d804c9b56b28a-64.webp","size":24}] 2d:["$","b",null,{"children":"Engram"}] 2e:["$","small",null,{"children":"#650"}] 22:o11,20:false 1f:7027 22:o17,1e:C 1d:C 1:C 15:C b:C 22:C </article></body></html>

22:o1,#22:o643,2:"$Sreact.fragment" 3:"/_next/static/immutable/chunks/12yk9g49l7-7r.js" 4:"/_next/static/immutable/chunks/2jrhxdtf12-uf.js" 5:I[7575,["$3","$4"],""] 6:I[10291,["$3"],"default"] 7:I[2968,["$3"],"default"] 8:I[4098,["$3"],"WebMCP"] 9:I[57507,["$3"],"ExternalLinks"] a:I[93137,["$3"],"Analytics"] 10:"$Sreact.suspense" 16:"ViewportBoundary" 17:I[82520,["$3"],"$16"] 19:"MetadataBoundary" 1a:I[82520,["$3"],"$19"] :HL["/_next/static/immutable/chunks/3g2_9hnyt6p64.css","style"] 1:X b:X d:X 15:X f:[["children",{"s":"__PAGE__","h":49314,"d":{"r":["$","$2","c",{"children":[["$","$10",null,{"fallback":["$","div",null,{"className":"wrap","style":{"minHeight":"60vh"}}],"children":"$L11"}],["$L12","$L13"],"$L14"]}],"p":false,"v":"$15"}}]] e:[["children",{"s":{"n":"slug","t":"d","k":"arcgram","s":[]},"h":49250,"d":{"r":["$","$2","c",{"children":[null,["$","$L6",null,{"parallelRouterKey":"children","error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L7",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":"$undefined","forbidden":"$undefined","unauthorized":"$undefined"}]]}],"p":false,"v":"$d"},"c":"$Qf"}]] c:[["children",{"s":"connectors","h":49250,"d":{"r":["$","$2","c",{"children":[null,["$","$L6",null,{"parallelRouterKey":"children","error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L7",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":"$undefined","forbidden":"$undefined","unauthorized":"$undefined"}]]}],"p":false,"v":"$d"},"c":"$Qe"}]] 1d:X 1e:X 22:o1232,0:{"q":"","i":false,"S":true,"r":"$1","t":{"t":{"s":"","h":49234,"d":{"r":["$","$2","c",{"children":[[["$","link","0",{"rel":"stylesheet","href":"/_next/static/immutable/chunks/3g2_9hnyt6p64.css","precedence":"next","crossOrigin":"$undefined","nonce":"$undefined"}],["$","script","script-0",{"src":"/_next/static/immutable/chunks/12yk9g49l7-7r.js","async":true,"nonce":"$undefined"}]],["$","html",null,{"lang":"en","className":"ec5Qua_variable","children":["$","body",null,{"children":[["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@graph\":[{\"@type\":\"Organization\",\"@id\":\"https://agentready.co/#org\",\"name\":\"Agent Ready\",\"url\":\"https://agentready.co/\",\"logo\":\"https://agentready.co/icon-512.png\"},{\"@type\":\"WebSite\",\"@id\":\"https://agentready.co/#site\",\"name\":\"Agent Ready\",\"url\":\"https://agentready.co/\",\"publisher\":{\"@id\":\"https://agentready.co/#org\"}}]}"}}],["$","header",null,{"className":"site","children":["$","div",null,{"className":"wrap","children":[["$","$L5",null,{"href":"/","className":"brand","children":[["$","svg",null,{"className":"brand-mark","viewBox":"0 0 64 64","aria-hidden":"true","children":[["$","defs",null,{"children":["$","linearGradient",null,{"id":"arg","x1":"0","y1":"1","x2":"1","y2":"0","children":[["$","stop",null,{"offset":"0","stopColor":"#2563EB"}],["$","stop",null,{"offset":"1","stopColor":"#06B6D4"}]]}]}],["$","path",null,{"d":"M9 55 L32 9 L55 55","fill":"none","stroke":"url(#arg)","strokeWidth":"11","strokeLinecap":"round","strokeLinejoin":"round"}],["$","circle",null,{"cx":"32","cy":"45","r":"7","fill":"#FACC15"}]]}],"Agent Ready"]}],["$","nav",null,{"aria-label":"Main","children":[["$","$L5",null,{"href":"/","children":"Connectors"}],["$","$L5",null,{"href":"/categories","className":"nv-x","children":"Categories"}],["$","$L5",null,{"href":"/blog","children":"Blog"}],["$","$L5",null,{"href":"/mcp-plugin-usage","className":"nv-x","children":"Usage"}],["$","$L5",null,{"href":"/methodology","className":"nv-x","children":"Methodology"}],["$","$L5",null,{"href":"/work-with-us","className":"nv-cta","children":"Work with us"}]]}]]}]}],["$","main",null,{"children":["$","$L6",null,{"parallelRouterKey":"children","error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L7",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[[["$","title",null,{"children":"404: This page could not be found."}],["$","div",null,{"style":{"fontFamily":"system-ui,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif,\"Apple Color Emoji\",\"Segoe UI Emoji\"","height":"100vh","textAlign":"center","display":"flex","flexDirection":"column","alignItems":"center","justifyContent":"center"},"children":["$","div",null,{"children":[["$","style",null,{"dangerouslySetInnerHTML":{"__html":"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}"}}],["$","h1",null,{"className":"next-error-h1","style":{"display":"inline-block","margin":"0 20px 0 0","padding":"0 23px 0 0","fontSize":24,"fontWeight":500,"verticalAlign":"top","lineHeight":"49px"},"children":404}],["$","div",null,{"style":{"display":"inline-block"},"children":["$","h2",null,{"style":{"fontSize":14,"fontWeight":400,"lineHeight":"49px","margin":0},"children":"This page could not be found."}]}]]}]}]],[]],"forbidden":"$undefined","unauthorized":"$undefined"}]}],["$","$L8",null,{}],["$","$L9",null,{}],["$","$La",null,{}],["$","footer",null,{"className":"foot","children":["$","div",null,{"className":"wrap","children":[["$","span",null,{"children":"Agent Ready"}],["$","nav",null,{"aria-label":"Footer","children":[["$","$L5",null,{"href":"/blog","children":"Blog"}],["$","$L5",null,{"href":"/mcp-plugin-usage","children":"Usage"}],["$","$L5",null,{"href":"/methodology","children":"Methodology"}],["$","$L5",null,{"href":"/categories","children":"Categories"}],["$","a",null,{"href":"/llms.txt","children":"llms.txt"}],["$","a",null,{"href":"/sitemap.xml","children":"Sitemap"}],["$","$L5",null,{"href":"/work-with-us","children":"Work with us"}]]}]]}]}]]}]}]]}],"p":false,"v":"$b"},"c":"$Qc"},"h":{"r":["$","$2","h",{"children":[null,["$","$L17","9B7-Qb6v-0RqjYUSCigMEv",{"children":"$L18"}],["$","$L1a","9B7-Qb6v-0RqjYUSCigMEm",{"children":[["$","div",null,{"hidden":true,"children":["$","$10",null,{"name":"Next.Metadata","children":"$L1b"}]}],"$L1c"]}]]}],"p":false,"v":"$1d"}},"s":"$1e","a":"$@1f","w":"$@20"} 22:o1b1,21:I[82520,["$3"],"OutletBoundary"] :HL["/_next/static/immutable/chunks/2lo4itc7piqdu.css","style"] d:C 12:["$","link","0",{"rel":"stylesheet","href":"/_next/static/immutable/chunks/2lo4itc7piqdu.css","precedence":"next","crossOrigin":"$undefined","nonce":"$undefined"}] 13:["$","script","script-0",{"src":"/_next/static/immutable/chunks/2jrhxdtf12-uf.js","async":true,"nonce":"$undefined"}] 14:["$","$L21",null,{"children":"$@22"}] 22:o14d,1e:300 18:[["$","meta","0",{"charSet":"utf-8"}],["$","meta","1",{"name":"viewport","content":"width=device-width, initial-scale=1"}],["$","meta","2",{"name":"theme-color","content":"#ffffff","media":"(prefers-color-scheme: light)"}],["$","meta","3",{"name":"theme-color","content":"#0d0d0f","media":"(prefers-color-scheme: dark)"}]] 22:o25,1d:0 15:0 15:T4,slug1d:T4,slug1e:300 22:o88,23:I[37521,["$3","$4"],"Logo"] 24:I[53913,["$3","$4"],"About"] 25:I[53913,["$3","$4"],"Explore"] 2b:I[60329,["$3"],"IconMark"] 26:T3a12,22:o3a12,--- name: arcgram description: "Generate an interactive node-and-edge diagram as a single standalone HTML file (Arcgram v2). Triggers on: arcgram, interactive/architectural diagram, topology, node graph, flow diagram, system map, dependency graph, workflow diagram, knowledge map, 'show me how X connects to Y', 'visualize this workflow', 'make this a topology', 'build a diagram of this'. Prefer for any system map, dependency graph, topology, or workflow picture even if interactivity wasn't asked for; on a follow-up 'make it inline' / 'show me' / 'render in chat', stay within Arcgram (Mode 1/2), don't switch tools." license: Apache-2.0 --- # Arcgram — Interactive Architectural Diagram (v2) > **For any AI agent — Arcgram is agent-neutral, not a Claude-only skill.** The block above is ordinary YAML frontmatter: Claude's skill system reads `name`/`description` to auto-trigger it, and any other agent (GPT · Gemini · Cline · Continue.dev · Cursor · Aider · local) just reads the plain-markdown body — no rejection, no Claude dependency. This is the **v2** skill — engine `template-v2.html`, fields in `schema.md`, per-platform install in `USAGE.md`. ## What this is Arcgram is a human↔AI collaboration substrate: the AI externalizes its reasoning as a typed node-and-edge graph the human audits and adjusts, converging on ground truth that survives across turns, sessions, and agents. **The AI draws, the human audits.** (Full positioning: `README.md`.) **Three guards keep the drawing honest** — internalize them before you build: - A **node** is a diagram element — a component, step, decision, artifact, or tool. It is **not** an HTML/DOM/React node, and **not** an argument-map "claim." An **edge** carries flow or dependency — **not** "supports/contradicts." - **The gap signal:** if a step can't be drawn as explicit nodes and edges, that's reasoning with a gap to expose — not a formality to skip. - **Drawing exposes gaps; it does not certify correctness.** A graph you can draw can still be wrong. --- ## When to use Reach for Arcgram when the structure of a thing — or of your reasoning about it — gets hard to hold in plain text: - a system map, dependency graph, topology, architecture, or data/work flow - "show me how X connects to Y" — relationships or lifecycles hard to track in prose - a decision / thinking flow (branches, forks) you want **checkable**, not just described - several people or agents need one shared, point-at-able source of truth ## Do NOT use for Charts (use a chart tool) · sequence diagrams (use a sequence-diagram tool) · mind maps · auto-layout graphs with hundreds of nodes (this skill assumes deliberate manual placement). ## Files in this skill bundle | File | Purpose | |---|---| | `template-v2.html` | Renderer engine + `DATA SECTION`. Copy it, fill the data near the top, re-open in a browser. One self-contained file — theme inlined. | | `schema.md` | **Full field reference — the single source** for every node / edge / band / column field. Read it; don't author a schema from memory. | | `layout-tips.md` | Positioning + edge-routing heuristics. | | `examples/` | Worked diagrams (system map · H bands · decision diamonds · workflow) — open one first. | | `skills/checkpoint/`, `reconcile/`, `validate/` | Headless validators — see step 7. | | `extensions/` | Optional overlays for a rendered flow — notably `arcgram-bugmarks.js`, the **Audit** defect-review UI. See § Audit. | | `themes/` | `base.css` + `default.css` — reference palette for forking. Optional at runtime: an export inlines the active theme, so the exported diagram needs no themes folder. | | `new-flow.mjs` | Scaffold a blank flow into `output/`: `node new-flow.mjs my-flow`. | | `output/` | **Default home for the flows you generate.** Yours — not part of the release, not leak-scanned, not shipped. | --- ## Choose the layout first (pin before building — it's hard to undo) Pin the layout before you draw — later edits build on it. There are two decisions, and the agent should **derive a recommendation from the flow's logic and help the user confirm it** rather than leave them to guess: 1. **Shape.** Do the nodes fall into **parallel categories** (subsystems, tracks, layers that coexist and wire across each other)? **Yes → group them:** `COLUMNS` (vertical) or `BANDS` (horizontal). **No** (one causal/branching chain) → **free DAG:** leave `COLUMNS`/`BANDS` empty, each node's role on its `cat`. 2. **Decisions.** Does it branch on yes/no forks? **Yes → add diamonds** (`kind:'diamond'`) — an overlay on *any* shape, not a separate option. See `schema.md § Decision diamond`. | shape ↓ · decisions → | plain | + diamonds | |---|---|---| | parallel categories · vertical | V — `COLUMNS` | V + thinking | | parallel categories · horizontal | H — `BANDS` | H + thinking | | no categories (logic chain) | free DAG | free thinking-flow | **Orientation (V vs H)** applies only when grouped and is the user's preference — so the agent proposes shape + decisions and **asks the user the reading direction** (default V): top→bottom story (lifecycles, pipelines) → `COLUMNS`; left→right stages/tracks → `BANDS` (`layout-tips.md §10b`). If the platform has a selection UI (`AskUserQuestion`), use it; otherwise ask inline and wait. ⚠ Never force a no-category chain into `COLUMNS`/`BANDS` — the tags float off their nodes. Keep the same mode on later edits unless the user asks to switch. --- ## Workflow > **Hard gate — read before you draw.** Before you place a single node, read `schema.md` (every field) and `layout-tips.md` (routing + positioning), and open one file in `examples/`. Authoring from memory of how generic diagram tools look — skipping these reads — is the single biggest cause of broken graphs: drifted field names, all-to-all edge meshes, and wires routed straight through nodes. Skipping the reads is the bug, not a shortcut. > > **Decide Audit up front.** Ask the user whether they want an **Audit** defect review of the finished flow (§ Audit). Settle it before you draw — never auto-annotate defects with icons. ### 1. Understand the topology Group nodes by category, decide reading direction, mark the critical connections. **Draw only the edges that exist in the real structure** — never wire every node in one group to every node in the next (a *mesh*). If a band-to-band connection comes out all-to-all, a structural element is missing (usually a convergence node — many inputs feed one decision); see `layout-tips.md §10b`. Logic-check the draft (step 7, pre-draw) before you place anything. ### 2. Read the references once `schema.md` for every field, `layout-tips.md` for positioning — read them instead of inlining a schema from memory. `schema.md` is gated against the engine (`schema-xref`), so any drift is caught. ### 3. Pin layout + sizing Pin the shape before drawing (§ Choose the layout first). The engine sizes nodes from their labels — you don't set `w`/`h` except on diamonds. Plan spacing per `schema.md § Sizing rules of thumb` + `layout-tips.md §3`. ### 4. Fill the DATA SECTION Scaffold a copy into `output/` — `node new-flow.mjs my-flow` → `output/my-flow.html` — (or copy `template-v2.html` by hand). Then in the `DATA SECTION` set ``, the title lines, `W`/`H`, and fill `nodes[]`, `edges[]`, and optionally `BANDS[]`, `COLUMNS{}` / `HCOLUMNS[]`, `CANVASES[]`, `STATUS_LEGEND{}`. Minimal shape (full fields: `schema.md § Node` / `§ Edge`): ```js // node — required: id, label, cat, type, desc, x, y { id:'FEED', label:'Feeding', cat:'CORE LOOP', type:'core', role:'action', desc:'Tap to feed.\nRestores hunger.', x:220, y:120 } // edge — required: f, t, fromPt, toPt { f:'FEED', t:'FOOD', fromPt:'bot', toPt:'top', route:'vhv', style:'solid' } ``` Declare node ownership on **both sides** — a node's `band`/`free` (plus `column`/`colFree` in H) and the band/column `members[]` must agree (gated A27/A28). The engine does the rest: bands auto-fit, lanes auto-grow, loose entry nodes wrap in a LIFECYCLE band. (`schema.md § Node ownership` · `§ Column ownership` · `§ Auto LIFECYCLE band` · `§ Canvas filter`.) ### 5. Features & styling — all optional, all in the references Add only what the flow needs; every field is defined in `schema.md`, every heuristic in `layout-tips.md`: - **Node types** — built-ins + the custom-type recipe: `§ Adding a custom type`. - **Edge routing** — `vhv`/`hvh`/`3leg` per edge; pick `fromPt`/`toPt` so lines exit/enter clean (arrowhead points at `toPt`): `§ Routers` + `layout-tips.md §6, §9`. - **Edge styling** — solid/dashed/bold + `crit`: `§ Edge styling decision matrix`. - **Critical paths** — `crit:true` + `lbl:'🔑N +verb'`, 3–7 per 30 edges: `layout-tips.md §8`. - **Decisions** — `kind:'diamond'` + `branch:'Y'/'N'`: `§ Decision diamond`. - **Status dots + author flags** — `status` (+ `STATUS_LEGEND`) and `flag` (a lightweight per-node "look here" note): `§ Status dot` · `§ Audit marker`. For a real **defect review** — bugs, logic holes, geometry defects like a wire through a node — use the **Audit** extension, not `flag` icons (§ Audit). ### 6. Render — three modes (fall back in order; don't skip ahead) - **Mode 1 — inline interactive widget.** Render `template-v2.html` via the host's HTML tool (e.g. `mcp__visualize__show_widget`). Pan/zoom/hover native — best for iteration. - **Mode 2 — inline static SVG (your own output, not a tool call).** If Mode 1 is unavailable, write an `<svg>…</svg>` block directly in your reply, using arcgram conventions (bands as rects, nodes border+dark bg, `vhv`/`hvh`/`3leg` edges, pills, crit red). Don't switch to a different diagram tool — the chat-native SVG routes *around* the failed tool. - **Mode 3 — standalone HTML export.** Write `template-v2.html` with your data to a file the user opens. Always available with filesystem access; use for handoff. ### 7. Self-check — run the validators (headless Node, under `skills/<tool>/`) The diagram is a *spec* — check it, don't eyeball the render. **Pre-draw** (as soon as `nodes`/`edges` exist) — Checkpoint catches dangling edges, orphans, a diamond missing a branch, duplicate ids: ```bash node skills/checkpoint/checkpoint.mjs your.html ``` **Post-draw** — Validate runs Checkpoint always, and Reconcile too when you pass matching code (a no-code run reports INCOMPLETE, never a silent pass): ```bash node skills/validate/validate.mjs your.html [code-file ...] ``` The engine also self-checks on render (stamps `self-check ran … | N findings` into the subtitle); the CLI is the enforce-by-construction version. ### 8. Save + syntax-check (Mode 3) Extract the inline JS and syntax-check before opening — catches typos that silently blank the canvas: ```bash awk '/<script>/{flag=1;next}/<\/script>/{flag=0}flag' your.html > /tmp/x.js && node --check /tmp/x.js ``` Open in browser. Confirm. --- ## Audit — defect review (opt-in; use the extension, not `flag` icons) Audit is a defect-review overlay for a *drawn* flow: a pulsing ring + a short reason tag on each marked node or edge, plus an "Audit" findings list. It **self-hides when there are no marks**, so a clean flow looks identical with or without it. **Ask first, mark only when asked.** Offer the user an Audit pass and settle it up front — do not auto-annotate defects. **Use the extension — not the `flag` field.** The Audit UI is `extensions/arcgram-bugmarks.js`. The per-node `flag` field draws a small icon but is *not* the Audit overlay — it must never stand in for a defect review. Load the extension *after* the engine (inline the `<script>` for a self-contained file), then call `ArcgramBugs.set([...])`: ```html <script src="extensions/arcgram-bugmarks.js"></script> <script> ArcgramBugs.set([ { id:'BUG-260911-ag-01', level:'high', type:'logic', anchor:{ node:'NODE_ID' }, reason:'no failure branch', desc:'a declined payment dead-ends here — the path assumes success', evidence:'scanner: 1 outgoing edge (success), 0 failure branch' }, { id:'BUG-260911-ag-02', level:'medium', type:'logic', anchor:{ edge:{ f:'A', t:'B' } }, reason:'unstated precondition', desc:'…', evidence:'doc: the spec requires a confirmed session here' } ]); </script> ``` `level` is **severity — how bad it is**, not where it came from: **`high`** = the flow is wrong and someone will hit it, **`medium`** = a real concern worth a decision, **`low`** = an advisory. (The checkpoint skill's `severity` field is a *different* measurement that happens to share the word — never map one onto the other.) Where a mark came from goes in `evidence`, prefixed `scanner:` / `doc:` / `derived:`. Mark **reasoning**, not geometry: use `type:'logic'`. Geometry is gated, not annotated — the gate bundle checks it and the engine fixes it. Anchor each mark to a `node:'ID'` or an `edge:{ f, t }`, give it an id shaped `BUG-<yymmdd>-<agent>-<seq>`, and ground it with `evidence`. Full API, the closure fields, and the id rules: `extensions/README.md`. **Closed marks on the flow — an authored per-map default.** Resolved rings are hidden by default (the header then says "N closed hidden"); a map whose closure evidence is part of the story declares `` before the extension's script region and its closed rings show from the first load. Settle it with the user like the Audit pass itself. Stamp semantics + the unknown-stamp allow-list: `extensions/README.md § Per-map stamps`. --- ## Hard rules Non-negotiables. Adjust everything else freely; these hold on every diagram. - **Write permission (two halves).** The agent edits **only the DATA SECTION** — everything below `END OF DATA SECTION` is the engine, off-limits. AND the agent **proposes; the human holds final write-authority** — draft and adjust the diagram data freely, but do not commit it as final shared ground truth without the human's confirmation. - **Author from `schema.md`, never from memory** of what generic diagram tools look like. Read this body + `schema.md` before generating; name fields, don't invent them. - **Single source.** `schema.md` owns every field; `README.md` owns the positioning. Point to them — don't restate a field definition or re-argue the job inside this file. - **Keep attribution.** The "Made with Arcgram" mark + `NOTICE` stay intact when you redistribute (Apache §4(d)); an unbranded / no-attribution build needs the commercial license (`WATERMARK-AND-COMMERCIAL-TERMS.md`). --- ## License **Apache-2.0.** Attribution rules → `§ Hard rules` (the canonical copy); full terms, trademark, and the unbranded commercial build → `LICENSE` · `NOTICE` · `WATERMARK-AND-COMMERCIAL-TERMS.md`. 22:o8,27:Td7e,22:od7e,--- name: checkpoint description: >- Validate an Arcgram diagram as a logic spec: is the flow well-formed and coherent? Reads node/edge structure (id/f/t/kind/role/type), never geometry. Typed findings - dangling edge, orphan node, missing branch, disconnected island, duplicate id, missing schema, plus optional coherence (every action traces to a goal). Marks, never blocks. Run after authoring or editing a diagram. Agent-neutral. Sibling to Reconcile (Checkpoint = structure; Reconcile = values). --- # Checkpoint - Arcgram logic self-check validator One input: a **diagram** (treated as a spec). One question: **is the flow logic well-formed and coherent?** Returns typed findings; marks, never blocks. Sibling to Reconcile: Checkpoint checks **structure** from the diagram alone; Reconcile checks **values** across diagram + code. Run both -> the diagram is a faithful, current spec. It checks the graph's logic structure - **not** drawing quality (the engine auto-handles overlaps/spacing/routing) and **not** semantic correctness (whether the logic is *right* - a human call). **Red line:** reads only the semantic field values `id`, `f`, `t`, `kind`, `role`, `type` (plus optional `status`, `flag`). Never a coordinate, color, size, or spacing. The judge: am I reading a semantic string, or a rendered visual quantity? ## Findings **Tier A - structural integrity (always on).** Domain-agnostic graph theory. | finding | means | |---|---| | `dangling-edge` | an edge points at an id that is not a node | | `orphan-node` | a node has no edge (chain broken) - except a declared `type:'open'` | | `missing-branch` | a decision (`kind:'diamond'`) lacks its Y or N branch | | `disconnected-island` | a cluster with no edge to the main flow | | `duplicate-id` | a node id used more than once | | `missing-field` | a node has no id, or an edge no `f`/`t` | | `missing-meta-version` | `ARCGRAM_META.version` absent | | `empty-diagram` | no nodes - reported uncovered, never falsely "clean" | **Tier B - justification coherence (auto-on for a thinking flow).** Does it read as reasoning: every `action` traces back to a `goal`, every `goal` reaches an `action`, every `decision` sits on a justification path. Auto-on when the diagram has a decision diamond (`kind:'diamond'`, = the engine's `_isThinkflow`); no-diamond diagrams skip it. Override `--tier-b` / `--no-tier-b`. Roles: `goal`/`action`/`decision` (coherence-bearing) + `constraint`/`evidence`/`state`/`output` (recognized). No resolvable role -> `[?]` uncovered, never guessed. `cat` is not read. Each finding is `{ id, type, severity, note }`; severity = `defect` or `uncovered`. ## Output (same contract as Reconcile) ``` { findings:[{id,type,severity,note}], summary, attestation, clean } self-check ran HH:MM:SS | N findings | M unchecked ``` **clean only when N=0 AND M=0** - "checked & clean" never looks like "never checked". Human reads the marker; agent reads the `type`. ## Use ```sh node checkpoint.mjs diagram.html [--tier-b | --no-tier-b] [--json] cat diagram.json | node checkpoint.mjs ``` ```js import { checkpoint } from './checkpoint.mjs'; checkpoint({ nodes, edges, meta }); ``` Exit 1 on any defect (advisory - never blocks). Verify: `node checkpoint-selftest.mjs` (19 checks). Render-time: the engine exposes the same check via `window.__arcgramInspect()` (the `checkpoint` field of its result) + the on-canvas `[?]` badge and attestation stamp. 22:o8,28:Td92,22:od92,--- name: reconcile description: >- Check that an Arcgram diagram's @spec values still match the code that implements it. Reads each node's desc "@spec key=value" line, finds the same-named code constant (hold_ms <-> HOLD_MS / holdMs), diffs the literals. Typed findings: drift, missing constant, non-literal, ambiguous. Reads only the @spec line + named constants, never geometry; marks, never blocks. Run after editing a diagram or its code. Agent-neutral. Sibling to Checkpoint (Checkpoint = structure; Reconcile = values). --- # Reconcile - code<->diagram @spec bijection validator Two inputs: a **diagram** + the **code** it describes. One question: **do the diagram's `@spec` values still equal the same-named code constants?** Returns typed findings; marks, never blocks. Sibling to Checkpoint: Checkpoint checks **structure** from the diagram alone; Reconcile checks **values** across diagram + code. Run both -> the diagram is a faithful, current spec. ## Input: the `@spec` line A node pins values on one line at the top of `desc`: ``` @spec hold_ms=800 fade_ms=200 <- Reconcile reads ONLY this line ``` - Prefix `@spec ` (ASCII). `key=value`, **key = the code constant name**. - Match is case/separator-insensitive: `hold_ms` = `HOLD_MS` = `holdMs` = `hold-ms`. - Value: a number (`800`, `0x320`) or a quoted string (`"left"`). - Free prose and a `~ tune: <range>` line are for humans - never read. Reconcile reads **only** the `@spec` line + **named** code constants. No geometry, no prose, no program behavior. ## Findings | finding | severity | means | |---|---|---| | `spec-drift` | defect | @spec value != the code constant | | `spec-no-constant` | defect | no same-named constant in the code | | `malformed-spec` | defect | @spec line is not `key=value` | | `value-uncovered` | uncovered | constant is computed, not a literal - no verdict | | `ambiguous-constant` | uncovered | two constants normalize to one name | | `no-spec-lines` | uncovered | nothing pinned - never a false "clean" | **Anchored on the `@spec` line** (not the code): drift is caught both ways, but a code constant on no `@spec` line is normal, not a finding. The extractor is a tolerant, dependency-free scan of `const/let/var =`, `NAME =`, `NAME:`, `#define`, `:=`, `[modifiers] const|readonly TYPE NAME =`, and C#/Unity **bare field initializers** `[attr] public|private|static TYPE NAME = value` (ScriptableObject tuning fields, no const/readonly) (JS/TS/Python/C/Go/C#/Java); numbers normalize (`800` == `800.0` == `0x320`, and a C#/Java numeric suffix is dropped so `0.2f` == `0.2`, `800L` == `800`). A Unity bare field is matched only with a field-only marker (an access modifier, `[Attribute]`, or `static`), so a method-body local (`int x = 5;`) is never captured. ## Output (same contract as Checkpoint) ``` { findings:[{id,type,severity,note}], summary, attestation, clean } self-check ran HH:MM:SS | N findings | M unchecked ``` **clean only when N=0 AND M=0** - "checked & clean" never looks like "never checked". Human reads the marker; agent reads the `type`. ## Use ```sh node reconcile.mjs diagram.html code.js [more.js ...] [--json] ``` ```js import { reconcile } from './reconcile.mjs'; reconcile({ nodes }, codeText); ``` Exit 1 on any defect (advisory - never blocks). Verify: `node reconcile-selftest.mjs` (48 checks). Example: `../../examples/example-workflow.html` vs `example-workflow.config.cs` (clean, reconciles 0 findings). 22:o9,29:T1073,22:o1073,--- name: validate description: >- Run both Arcgram self-checks in one call. A thin wrapper over Checkpoint (structure) and Reconcile (values): validate(diagram,{code}) runs Checkpoint always and Reconcile when code is supplied, then merges into one source-tagged result in the same {findings,summary,attestation,clean} contract. clean = checkpoint.clean && reconcile.clean. Adds no new check; marks, never blocks. Run after authoring or editing a diagram and its code. Agent-neutral. The single door over its two siblings. --- # Validate - the combined Arcgram self-check entry One call. Two questions answered between the siblings: **is the flow well-formed and coherent, AND do its `@spec` values still match the code?** Validate is a **thin wrapper**, not a new check. It imports [Checkpoint](../checkpoint/SKILL.md) and [Reconcile](../reconcile/SKILL.md) **untouched** and composes their pure functions: - **Checkpoint** (one input: the diagram) checks **structure** - well-formed, coherent. - **Reconcile** (two inputs: diagram + code) checks **values** - `@spec` literals == code constants. Run both -> the diagram is a faithful, current spec. Every finding still comes from one sibling; Validate only **merges, source-tags, and unions the attestation**. ## Code is optional | input | what runs | clean possible? | |---|---|---| | `validate(diagram, { code })` | Checkpoint **and** Reconcile | yes | | `validate(diagram)` (no code) | Checkpoint only | no - values unchecked | Without code there is nothing to reconcile, so Reconcile is **not run** (running it would just flag every `@spec` pin as missing). Instead the value half is reported as one `uncovered` `value-check-not-run` finding, so a structure-only run is never falsely "clean" on values - the same anti-silent-pass rule both siblings enforce. **Locked invariant - `uncovered` is a third state, never a pass.** A diagram-only run (no code) can **never** produce an overall green clean, even when the structure is perfect. The attestation explicitly appends `| code consistency: NOT CHECKED (no code supplied)`, `summary.codeConsistency` reads `"NOT_CHECKED"`, and the CLI verdict is `INCOMPLETE`, never the word CLEAN. This guards the exact silent-pass the family exists to prevent: claiming "checked & clean" while the code half was never checked. ## Output (same contract as both siblings, plus `source`) ``` { findings:[{id,type,severity,note,source}], summary, attestation, clean } self-check ran HH:MM:SS | N findings | M unchecked ``` - **`source`** is `"checkpoint"` | `"reconcile"` | `"validate"` - so you can route a finding back to the check that raised it. Everything else (`id`, `type`, `severity`, `note`) is verbatim from the sibling. - **`attestation`** is the **union**: N and M are the combined defect / uncovered totals. When the value half did not run it appends an explicit `| code consistency: NOT CHECKED (no code supplied)`. - **`clean === checkpoint.clean && reconcile.clean`** (identically `N=0 && M=0`); `uncovered` is a third state and is never folded into `clean`, so a skipped value-check is not clean. - **`summary`** nests each sibling's own summary under `.checkpoint` / `.reconcile`, plus `ran` and `codeConsistency` (`"checked"` | `"NOT_CHECKED"`). - The full sub-results are returned verbatim on `.checkpoint` / `.reconcile` for a caller that wants the per-check breakdown. Human reads the marker; agent reads the `type` (and `source`). ## Use ```sh node validate.mjs <diagram.html | diagram.json> [code-file ...] [--tier-b | --no-tier-b] [--json] ``` ```js import { validate } from './validate.mjs'; validate({ nodes, edges, meta }, { code }); // code: string | string[]; omit -> structure only ``` `--tier-b` / `--no-tier-b` forward to Checkpoint's justification tier (else auto-detected by a decision diamond). Exit 1 on any defect (advisory - never blocks; an uncovered value-check is not a defect). Verify: `node validate-selftest.mjs` (36 checks - source tags, union attestation, clean rule, the code-omitted path, and byte parity with running the two siblings separately). Example: `examples/example-workflow.html` vs `example-workflow.config.cs` (clean). 22:oc51,11:[["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@type\":\"SoftwareApplication\",\"name\":\"Arcgram\",\"description\":\"Surface an AI agent's thinking and make it auditable. Instead of a wall of prose, your agent commits its reasoning to typed nodes and explicit edges — real data you inspect and correct node by node, converging on ground truth that carries across turns, sessions, and agents. Use it as an agent harness or operating frame: propose → draw → self-audit → fix, human in the loop. Delivered as a single st\",\"applicationCategory\":\"AI assistant add-ons\",\"operatingSystem\":\"Claude\",\"url\":\"https://agentready.co/connectors/arcgram\",\"sameAs\":[],\"publisher\":{\"@type\":\"Organization\",\"name\":\"jovesun-lab\",\"url\":\"https://github.com\"}}"}}],["$","div",null,{"className":"wrap","children":[["$","div",null,{"className":"pp-head","children":[["$","$L23",null,{"name":"Arcgram","url":null,"size":56}],["$","div",null,{"children":[["$","div",null,{"className":"crumbs","children":[["$","$L5",null,{"href":"/","children":"Connectors"}],[" / ",["$","$L5",null,{"href":"/categories/ai-tools","children":"AI assistant add-ons"}]]]}],["$","h1",null,{"children":["Arcgram",null]}]]}]]}],["$","div",null,{"className":"pp","children":[["$","div",null,{"className":"ca","children":[["$","p",null,{"className":"tag","children":"Surface an AI agent's thinking and make it auditable."}],["$","$L24",null,{"text":"Instead of a wall of prose, your agent commits its reasoning to typed nodes and explicit edges — real data you inspect and correct node by node, converging on ground truth that carries across turns, sessions, and agents. Use it as an agent harness or operating frame: propose → draw → self-audit → fix, human in the loop. Delivered as a single standalone interactive HTML file — no build step, no npm, no CDN, no runtime; open in any browser."}],["$","$L25",null,{"examples":[],"tools":[],"prefix":"","server":null,"instructions":[],"skills":[{"name":"arcgram","label":"arcgram","gist":"Generate an interactive node-and-edge diagram as a single standalone HTML file (Arcgram v2).","source":"https://github.com/jovesun-lab/arcgram/blob/fe4fc0eb7d0c6039e2051d02929e28ae5fd12252/SKILL.md","raw":"$26"},{"name":"checkpoint","label":"checkpoint","gist":"Validate an Arcgram diagram as a logic spec: is the flow well-formed and coherent?","source":"https://github.com/jovesun-lab/arcgram/blob/fe4fc0eb7d0c6039e2051d02929e28ae5fd12252/skills/checkpoint/SKILL.md","raw":"$27"},{"name":"reconcile","label":"reconcile","gist":"Check that an Arcgram diagram's @spec values still match the code that implements it.","source":"https://github.com/jovesun-lab/arcgram/blob/fe4fc0eb7d0c6039e2051d02929e28ae5fd12252/skills/reconcile/SKILL.md","raw":"$28"},{"name":"validate","label":"validate","gist":"Run both Arcgram self-checks in one call.","source":"https://github.com/jovesun-lab/arcgram/blob/fe4fc0eb7d0c6039e2051d02929e28ae5fd12252/skills/validate/SKILL.md","raw":"$29"}],"skillNote":"4 in the Claude listing"}],null]}],"$L2a"]}]]}]] 22:oa3d,1b:[["$","title","0",{"children":"Arcgram by jovesun-lab: AI connector · Agent Ready"}],["$","meta","1",{"name":"description","content":"Surface an AI agent's thinking and make it auditable. Instead of a wall of prose, your agent commits its reasoning to typed nodes and explicit edges — real…"}],["$","link","2",{"rel":"manifest","href":"/manifest.webmanifest","crossOrigin":"$undefined"}],["$","link","3",{"rel":"canonical","href":"https://agentready.co/connectors/arcgram"}],["$","meta","4",{"property":"og:title","content":"Arcgram by jovesun-lab: AI connector"}],["$","meta","5",{"property":"og:description","content":"Surface an AI agent's thinking and make it auditable. Instead of a wall of prose, your agent commits its reasoning to typed nodes and explicit edges — real…"}],["$","meta","6",{"property":"og:url","content":"https://agentready.co/connectors/arcgram"}],["$","meta","7",{"property":"og:site_name","content":"Agent Ready"}],["$","meta","8",{"property":"og:image","content":"https://agentready.co/connectors/arcgram/opengraph-image?3b61e23efdbb5225"}],["$","meta","9",{"property":"og:image:type","content":"image/png"}],["$","meta","10",{"property":"og:image:width","content":"1200"}],["$","meta","11",{"property":"og:image:height","content":"630"}],["$","meta","12",{"property":"og:image:alt","content":"AI connector on Agent Ready"}],["$","meta","13",{"property":"og:type","content":"website"}],["$","meta","14",{"name":"twitter:card","content":"summary_large_image"}],["$","meta","15",{"name":"twitter:title","content":"Arcgram by jovesun-lab: AI connector"}],["$","meta","16",{"name":"twitter:description","content":"Surface an AI agent's thinking and make it auditable. Instead of a wall of prose, your agent commits its reasoning to typed nodes and explicit edges — real…"}],["$","meta","17",{"name":"twitter:image","content":"https://agentready.co/connectors/arcgram/opengraph-image?3b61e23efdbb5225"}],["$","meta","18",{"name":"twitter:image:alt","content":"AI connector on Agent Ready"}],["$","meta","19",{"name":"twitter:image:type","content":"image/png"}],["$","meta","20",{"name":"twitter:image:width","content":"1200"}],["$","meta","21",{"name":"twitter:image:height","content":"630"}],["$","link","22",{"rel":"icon","href":"/favicon.ico?favicon.3iymezf4ycw3e.ico","sizes":"64x64","type":"image/x-icon"}],["$","link","23",{"rel":"icon","href":"/icon.svg?icon.1m07wgr9jqhm5.svg","sizes":"any","type":"image/svg+xml"}],["$","link","24",{"rel":"apple-touch-icon","href":"/apple-icon.png?apple-icon.3q8qrgufoflqu.png","sizes":"180x180","type":"image/png"}],["$","$L2b","25",{}]] 22:o10,1c:null 22:null 22:o22,:HL["/stores/claude.png","image"] 22:o128b,2a:["$","aside",null,{"className":"side","children":[["$","div",null,{"className":"card","children":[["$","h2",null,{"children":"Stores"}],["$","div",null,{"className":"st","children":[["$","div","claude",{"style":{"display":"contents"},"children":[["$","span",null,{"className":"s","children":[["$","img",null,{"className":"sm","src":"/stores/claude.png","alt":"","width":16,"height":16}],["$","b",null,{"children":"Claude"}],["$","a",null,{"className":"src","href":"https://github.com/jovesun-lab/arcgram","target":"_blank","rel":"noopener nofollow","children":"Source ↗"}]]}],["$","span",null,{"className":"rk2","children":[["$","small",null,{"children":"Listed, not ranked"}],null]}]]}]]}]]}],["$","div",null,{"className":"card","children":[["$","h2",null,{"children":"Details"}],["$","dl",null,{"className":"kv","children":[[["$","dt",null,{"children":"Company"}],["$","dd",null,{"children":"jovesun-lab"}]],[["$","dt",null,{"children":"Publisher"}],["$","dd",null,{"children":"Community"}]],["$","dt",null,{"children":"Category"}],["$","dd",null,{"children":["$","$L5",null,{"href":"/categories/ai-tools","children":"AI assistant add-ons"}]}],[["$","dt",null,{"children":"Audience"}],["$","dd",null,{"children":"For developers"}]],null,["$","dt",null,{"children":"Sign-in"}],["$","dd",null,{"children":"None needed"}],["$","dt",null,{"children":"Interactive UI"}],["$","dd",null,{"children":"No"}],[["$","dt",null,{"children":"Checked"}],["$","dd",null,{"children":"Oct 27, 2026"}]],null]}]]}],["$","div",null,{"className":"card","children":[["$","h2",null,{"children":["More in ","AI assistant add-ons"]}],["$","ul",null,{"className":"sim","children":[["$","li","14965",{"children":["$","$L5",null,{"href":"/connectors/rhythm-ai-personality-tuner","prefetch":false,"children":[["$","$L23",null,{"name":"Rhythm AI Personality Tuner","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/02/02846ae172d38bae727fc11c13349772f8d2840b740c604ba8c165bdd8d844a9-64.webp","size":24}],["$","b",null,{"children":"Rhythm AI Personality Tuner"}],["$","small",null,{"children":"#146"}]]}]}],["$","li","4302",{"children":["$","$L5",null,{"href":"/connectors/playmcp","prefetch":false,"children":[["$","$L23",null,{"name":"PlayMCP","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/7e/7e9da62ddba915cd96ea88590aa0dd924d416591cf546e27a8d6cb3fe821608b-64.webp","size":24}],["$","b",null,{"children":"PlayMCP"}],["$","small",null,{"children":"#152"}]]}]}],["$","li","15060",{"children":["$","$L5",null,{"href":"/connectors/prompt-perfect","prefetch":false,"children":[["$","$L23",null,{"name":"Prompt Perfect","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/25/25c8b7a6dc543f28b9684227604bf15cec842945f4ac99b7c7fc130d321a8a39-64.webp","size":24}],["$","b",null,{"children":"Prompt Perfect"}],["$","small",null,{"children":"#333"}]]}]}],["$","li","15246",{"children":["$","$L5",null,{"href":"/connectors/basic-memory-cloud","prefetch":false,"children":[["$","$L23",null,{"name":"Basic Memory Cloud","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/48/48765542121cf50a22435c94e535acdf031b1b4208770d51e61a2e1e619acc59-64.webp","size":24}],["$","b",null,{"children":"Basic Memory Cloud"}],["$","small",null,{"children":"#1,199"}]]}]}],["$","li","15582",{"children":["$","$L5",null,{"href":"/connectors/tasknotify","prefetch":false,"children":[["$","$L23",null,{"name":"TaskNotify","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/cc/ccef65bdab248018fd966ac128ac1efec782e1b012f181200f01cb5d4b04753a-64.webp","size":24}],["$","b",null,{"children":"TaskNotify"}],["$","small",null,{"children":"#476"}]]}]}],["$","li","15388",{"children":["$","$L5",null,{"href":"/connectors/app-atlas","prefetch":false,"children":[["$","$L23",null,{"name":"App Atlas","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/8f/8fce6de54fe02b96dd3c7c81562f4faa5a36c4b92a47080da489b59a10709b46-64.webp","size":24}],["$","b",null,{"children":"App Atlas"}],["$","small",null,{"children":"#560"}]]}]}],["$","li","15337",{"children":["$","$L5",null,{"href":"/connectors/caveman-mode","prefetch":false,"children":[["$","$L23",null,{"name":"Caveman Mode","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/27/27efe08ae01ebda82f25893226e33a2c9c3d49b6c6ec787b9a0bff471a580795-64.webp","size":24}],["$","b",null,{"children":"Caveman Mode"}],["$","small",null,{"children":"#604"}]]}]}],["$","li","15406",{"children":["$","$L5",null,{"href":"/connectors/engram-2","prefetch":false,"children":["$L2c","$L2d","$L2e"]}]}]]}]]}]]}] 22:o129,2c:["$","$L23",null,{"name":"Engram","url":"https://bodgzvrssuydrdmsrqsh.supabase.co/storage/v1/object/public/listing-media/v/37/37541cf9765efdad0155df1c3af9d933e5cd8be560b306e7281d804c9b56b28a-64.webp","size":24}] 2d:["$","b",null,{"children":"Engram"}] 2e:["$","small",null,{"children":"#650"}] 22:o11,20:false 1f:7027 22:o17,1e:C 1d:C 1:C 15:C b:C 22:C </article></body></html>

Arcgram

Skills

Arcgram

Skills