Skip to main content

orchestrator-show-progress-graph

Shows the orchestrator roadmap progress as a drawing (features per stage, dependencies, what is done, in progress, to do, blocked and what is stuck), as plain text for chat and e-mail, Mermaid for PR/GitHub and a short line for short-message channels. Writes .specs/project/GRAPH.md. Use in every checkpoint report, when the user asks "how is the loop going", "what is blocking", "show the graph/progress", or when assembling the loop notification and PR. Do NOT activate for code, database or architecture graphs.

Quellinformationen

Repository
iagopadilha/loop-engineering-harness
Letzte Quellaktivität
15. Oktober 2026 um 04:42
Erkannte Sprache von SKILL.md
Englisch
Sterne
0
Forks
0

Installationsoptionen

Standardmäßig ist der Prompt ausgewählt, der zuerst die Quelle prüft. Sie können zu einem direkten Befehl wechseln oder eine lokale Kopie herunterladen.

Quelldateien prüfen

Lesen Sie SKILL.md und alle von SkillsMP angezeigten Begleitdateien, bevor Sie sich für eine Installation entscheiden.

SKILL.md wird angezeigt

SKILL.md
Quellanweisungen · Schreibgeschützte Vorschau
name
orchestrator-show-progress-graph
description
Shows the orchestrator roadmap progress as a drawing (features per stage, dependencies, what is done, in progress, to do, blocked and what is stuck), as plain text for chat and e-mail, Mermaid for PR/GitHub and a short line for short-message channels. Writes .specs/project/GRAPH.md. Use in every checkpoint report, when the user asks "how is the loop going", "what is blocking", "show the graph/progress", or when assembling the loop notification and PR. Do NOT activate for code, database or architecture graphs.
# Show the progress graph Depends on the `orchestrator` skill (script `${CLAUDE_PLUGIN_ROOT}/skills/orchestrator/loop/graph_view.py`; reads `.specs/project/ROADMAP.md`). It only reads and draws: it changes nothing except `GRAPH.md`. ## Usage ``` python ${CLAUDE_PLUGIN_ROOT}/skills/orchestrator/loop/graph_view.py <root> # plain text (default) python ${CLAUDE_PLUGIN_ROOT}/skills/orchestrator/loop/graph_view.py <root> --short # one line per stage python ${CLAUDE_PLUGIN_ROOT}/skills/orchestrator/loop/graph_view.py <root> --mermaid # for PR / .md python ${CLAUDE_PLUGIN_ROOT}/skills/orchestrator/loop/graph_view.py <root> --write # writes .specs/project/GRAPH.md ``` Example output: ``` Stage 1 (2/3 closed) [x] api done [x] ui done [!] coupon BLOCKED (2 failures) Stage 2 (0/2 closed) [>] screen building <- api(done) ui(done) [.] end pending <- screen(building) STUCK on: coupon Legend: [x] done [>] in progress [.] to do [!] blocked (<- waits for) ``` (The output labels are the script's own: `Stage`, `closed`, `done`, `BLOCKED`, `building`, `pending`, `STUCK on`, `Legend`, `to do`, `in progress`, `waits for`; in the short line, stages are prefixed `S<n>`.) ## Where each format is used (already wired) | Where | Format | |---|---| | Checkpoint report in chat | plain text | | Loop e-mail notification | plain text in a block | | Short-message notification | `--short` (fits the ~1000-character limit) | | Delivery PR body and `loop-summary-*.md` | Mermaid | | `.specs/project/GRAPH.md` | text + Mermaid, rewritten by the script every round | ## When the orchestrator uses it - **Every checkpoint** (Phase 8) and before asking "shall I continue?": paste the plain text into the report. - When answering "how is it going?": run it and show it; do not describe from memory. - `STUCK on:` indicates a feature that depends on a blocked one: take it to the user.
Auf GitHub ansehen