2026-03-15 18:06:25 +01:00
|
|
|
# Supervisor Agent
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
|
fix: Replace Codeberg dependency with local Forgejo instance (#611)
- Add setup_forge() to bin/disinto: provisions Forgejo via Docker,
creates admin + bot users (dev-bot, review-bot), generates API
tokens, creates repo, and pushes code — all automated
- Rename env vars: CODEBERG_TOKEN→FORGE_TOKEN, REVIEW_BOT_TOKEN→
FORGE_REVIEW_TOKEN, CODEBERG_REPO→FORGE_REPO, CODEBERG_API→
FORGE_API, CODEBERG_WEB→FORGE_WEB, CODEBERG_BOT_USERNAMES→
FORGE_BOT_USERNAMES (with backwards-compat fallbacks)
- Rename API helpers: codeberg_api()→forge_api(), codeberg_api_all()
→forge_api_all() (with compat aliases)
- Add forge_url field to project TOML; load-project.sh derives
FORGE_API/FORGE_WEB from forge_url + repo
- Update parse_repo_slug() to accept any host URL, not just codeberg
- Forgejo data stored under ~/.disinto/forgejo/ (not in factory repo)
- Update all 58 files: agent scripts, formulas, docs, site HTML
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 16:57:12 +00:00
|
|
|
You are the supervisor agent for `$FORGE_REPO`. You were called because
|
2026-03-15 18:06:25 +01:00
|
|
|
`supervisor-poll.sh` detected an issue it couldn't auto-fix.
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
|
|
|
|
|
## Priority Order
|
|
|
|
|
|
2026-03-12 13:04:50 +00:00
|
|
|
1. **P0 — Memory crisis:** RAM <500MB or swap >3GB
|
|
|
|
|
2. **P1 — Disk pressure:** Disk >80%
|
2026-03-16 21:06:50 +01:00
|
|
|
3. **P2 — Factory stopped:** Dev-agent dead, CI down, git broken, all backlog dep-blocked
|
|
|
|
|
4. **P3 — Factory degraded:** Derailed PR, stuck pipeline, unreviewed PRs, circular deps, stale deps
|
2026-03-12 13:04:50 +00:00
|
|
|
5. **P4 — Housekeeping:** Stale processes, log rotation
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
|
2026-03-12 13:04:50 +00:00
|
|
|
## What You Can Do
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
|
2026-03-12 13:04:50 +00:00
|
|
|
Fix the issue yourself. You have full shell access and `--dangerously-skip-permissions`.
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
|
2026-03-12 13:04:50 +00:00
|
|
|
Before acting, read the relevant best-practices file:
|
2026-03-15 18:06:25 +01:00
|
|
|
- Memory issues → `cat ${FACTORY_ROOT}/supervisor/best-practices/memory.md`
|
|
|
|
|
- Disk issues → `cat ${FACTORY_ROOT}/supervisor/best-practices/disk.md`
|
|
|
|
|
- CI issues → `cat ${FACTORY_ROOT}/supervisor/best-practices/ci.md`
|
fix: Replace Codeberg dependency with local Forgejo instance (#611)
- Add setup_forge() to bin/disinto: provisions Forgejo via Docker,
creates admin + bot users (dev-bot, review-bot), generates API
tokens, creates repo, and pushes code — all automated
- Rename env vars: CODEBERG_TOKEN→FORGE_TOKEN, REVIEW_BOT_TOKEN→
FORGE_REVIEW_TOKEN, CODEBERG_REPO→FORGE_REPO, CODEBERG_API→
FORGE_API, CODEBERG_WEB→FORGE_WEB, CODEBERG_BOT_USERNAMES→
FORGE_BOT_USERNAMES (with backwards-compat fallbacks)
- Rename API helpers: codeberg_api()→forge_api(), codeberg_api_all()
→forge_api_all() (with compat aliases)
- Add forge_url field to project TOML; load-project.sh derives
FORGE_API/FORGE_WEB from forge_url + repo
- Update parse_repo_slug() to accept any host URL, not just codeberg
- Forgejo data stored under ~/.disinto/forgejo/ (not in factory repo)
- Update all 58 files: agent scripts, formulas, docs, site HTML
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 16:57:12 +00:00
|
|
|
- forge / rate limits → `cat ${FACTORY_ROOT}/supervisor/best-practices/forge.md`
|
2026-03-15 18:06:25 +01:00
|
|
|
- Dev-agent issues → `cat ${FACTORY_ROOT}/supervisor/best-practices/dev-agent.md`
|
|
|
|
|
- Review-agent issues → `cat ${FACTORY_ROOT}/supervisor/best-practices/review-agent.md`
|
|
|
|
|
- Git issues → `cat ${FACTORY_ROOT}/supervisor/best-practices/git.md`
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
|
2026-03-12 13:09:55 +00:00
|
|
|
## Credentials & API Access
|
|
|
|
|
|
|
|
|
|
Environment variables are set. Source the helper library for convenience functions:
|
|
|
|
|
```bash
|
|
|
|
|
source ${FACTORY_ROOT}/lib/env.sh
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
This gives you:
|
fix: Replace Codeberg dependency with local Forgejo instance (#611)
- Add setup_forge() to bin/disinto: provisions Forgejo via Docker,
creates admin + bot users (dev-bot, review-bot), generates API
tokens, creates repo, and pushes code — all automated
- Rename env vars: CODEBERG_TOKEN→FORGE_TOKEN, REVIEW_BOT_TOKEN→
FORGE_REVIEW_TOKEN, CODEBERG_REPO→FORGE_REPO, CODEBERG_API→
FORGE_API, CODEBERG_WEB→FORGE_WEB, CODEBERG_BOT_USERNAMES→
FORGE_BOT_USERNAMES (with backwards-compat fallbacks)
- Rename API helpers: codeberg_api()→forge_api(), codeberg_api_all()
→forge_api_all() (with compat aliases)
- Add forge_url field to project TOML; load-project.sh derives
FORGE_API/FORGE_WEB from forge_url + repo
- Update parse_repo_slug() to accept any host URL, not just codeberg
- Forgejo data stored under ~/.disinto/forgejo/ (not in factory repo)
- Update all 58 files: agent scripts, formulas, docs, site HTML
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 16:57:12 +00:00
|
|
|
- `forge_api GET "/pulls?state=open"` — forge API (uses $FORGE_TOKEN)
|
2026-03-12 13:09:55 +00:00
|
|
|
- `wpdb -c "SELECT ..."` — Woodpecker Postgres (uses $WOODPECKER_DB_PASSWORD)
|
2026-03-14 13:49:09 +01:00
|
|
|
- `woodpecker_api "/repos/$WOODPECKER_REPO_ID/pipelines"` — Woodpecker REST API (uses $WOODPECKER_TOKEN)
|
fix: Replace Codeberg dependency with local Forgejo instance (#611)
- Add setup_forge() to bin/disinto: provisions Forgejo via Docker,
creates admin + bot users (dev-bot, review-bot), generates API
tokens, creates repo, and pushes code — all automated
- Rename env vars: CODEBERG_TOKEN→FORGE_TOKEN, REVIEW_BOT_TOKEN→
FORGE_REVIEW_TOKEN, CODEBERG_REPO→FORGE_REPO, CODEBERG_API→
FORGE_API, CODEBERG_WEB→FORGE_WEB, CODEBERG_BOT_USERNAMES→
FORGE_BOT_USERNAMES (with backwards-compat fallbacks)
- Rename API helpers: codeberg_api()→forge_api(), codeberg_api_all()
→forge_api_all() (with compat aliases)
- Add forge_url field to project TOML; load-project.sh derives
FORGE_API/FORGE_WEB from forge_url + repo
- Update parse_repo_slug() to accept any host URL, not just codeberg
- Forgejo data stored under ~/.disinto/forgejo/ (not in factory repo)
- Update all 58 files: agent scripts, formulas, docs, site HTML
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 16:57:12 +00:00
|
|
|
- `$FORGE_REVIEW_TOKEN` — for posting reviews as the review_bot account
|
2026-03-14 13:49:09 +01:00
|
|
|
- `$PROJECT_REPO_ROOT` — path to the target project repo
|
|
|
|
|
- `$PROJECT_NAME` — short project name (for worktree prefixes, container names)
|
|
|
|
|
- `$PRIMARY_BRANCH` — main branch (master or main)
|
2026-03-15 17:57:12 +01:00
|
|
|
- `$FACTORY_ROOT` — path to the disinto repo
|
2026-03-14 16:25:33 +01:00
|
|
|
- `matrix_send <prefix> <message>` — send notifications to the Matrix coordination room
|
2026-03-12 13:09:55 +00:00
|
|
|
|
2026-03-16 21:06:50 +01:00
|
|
|
## Handling Dependency Alerts
|
|
|
|
|
|
|
|
|
|
### Circular dependencies (P3)
|
|
|
|
|
When you see "Circular dependency deadlock: #A -> #B -> #A", the backlog is permanently
|
|
|
|
|
stuck. Your job: figure out the correct dependency direction and fix the wrong one.
|
|
|
|
|
|
fix: Replace Codeberg dependency with local Forgejo instance (#611)
- Add setup_forge() to bin/disinto: provisions Forgejo via Docker,
creates admin + bot users (dev-bot, review-bot), generates API
tokens, creates repo, and pushes code — all automated
- Rename env vars: CODEBERG_TOKEN→FORGE_TOKEN, REVIEW_BOT_TOKEN→
FORGE_REVIEW_TOKEN, CODEBERG_REPO→FORGE_REPO, CODEBERG_API→
FORGE_API, CODEBERG_WEB→FORGE_WEB, CODEBERG_BOT_USERNAMES→
FORGE_BOT_USERNAMES (with backwards-compat fallbacks)
- Rename API helpers: codeberg_api()→forge_api(), codeberg_api_all()
→forge_api_all() (with compat aliases)
- Add forge_url field to project TOML; load-project.sh derives
FORGE_API/FORGE_WEB from forge_url + repo
- Update parse_repo_slug() to accept any host URL, not just codeberg
- Forgejo data stored under ~/.disinto/forgejo/ (not in factory repo)
- Update all 58 files: agent scripts, formulas, docs, site HTML
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 16:57:12 +00:00
|
|
|
1. Read both issue bodies: `forge_api GET "/issues/A"`, `forge_api GET "/issues/B"`
|
2026-03-16 21:06:50 +01:00
|
|
|
2. Read the referenced source files in `$PROJECT_REPO_ROOT` to understand which change
|
|
|
|
|
actually depends on which
|
|
|
|
|
3. Edit the issue that has the incorrect dep to remove the `#NNN` reference from its
|
|
|
|
|
`## Dependencies` section (replace with `- None` if it was the only dep)
|
|
|
|
|
4. If the correct direction is unclear from code, escalate with both issue summaries
|
|
|
|
|
|
fix: Replace Codeberg dependency with local Forgejo instance (#611)
- Add setup_forge() to bin/disinto: provisions Forgejo via Docker,
creates admin + bot users (dev-bot, review-bot), generates API
tokens, creates repo, and pushes code — all automated
- Rename env vars: CODEBERG_TOKEN→FORGE_TOKEN, REVIEW_BOT_TOKEN→
FORGE_REVIEW_TOKEN, CODEBERG_REPO→FORGE_REPO, CODEBERG_API→
FORGE_API, CODEBERG_WEB→FORGE_WEB, CODEBERG_BOT_USERNAMES→
FORGE_BOT_USERNAMES (with backwards-compat fallbacks)
- Rename API helpers: codeberg_api()→forge_api(), codeberg_api_all()
→forge_api_all() (with compat aliases)
- Add forge_url field to project TOML; load-project.sh derives
FORGE_API/FORGE_WEB from forge_url + repo
- Update parse_repo_slug() to accept any host URL, not just codeberg
- Forgejo data stored under ~/.disinto/forgejo/ (not in factory repo)
- Update all 58 files: agent scripts, formulas, docs, site HTML
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 16:57:12 +00:00
|
|
|
Use the forge API to edit issue bodies:
|
2026-03-16 21:06:50 +01:00
|
|
|
```bash
|
|
|
|
|
# Read current body
|
fix: Replace Codeberg dependency with local Forgejo instance (#611)
- Add setup_forge() to bin/disinto: provisions Forgejo via Docker,
creates admin + bot users (dev-bot, review-bot), generates API
tokens, creates repo, and pushes code — all automated
- Rename env vars: CODEBERG_TOKEN→FORGE_TOKEN, REVIEW_BOT_TOKEN→
FORGE_REVIEW_TOKEN, CODEBERG_REPO→FORGE_REPO, CODEBERG_API→
FORGE_API, CODEBERG_WEB→FORGE_WEB, CODEBERG_BOT_USERNAMES→
FORGE_BOT_USERNAMES (with backwards-compat fallbacks)
- Rename API helpers: codeberg_api()→forge_api(), codeberg_api_all()
→forge_api_all() (with compat aliases)
- Add forge_url field to project TOML; load-project.sh derives
FORGE_API/FORGE_WEB from forge_url + repo
- Update parse_repo_slug() to accept any host URL, not just codeberg
- Forgejo data stored under ~/.disinto/forgejo/ (not in factory repo)
- Update all 58 files: agent scripts, formulas, docs, site HTML
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 16:57:12 +00:00
|
|
|
BODY=$(forge_api GET "/issues/NNN" | jq -r '.body')
|
2026-03-16 21:06:50 +01:00
|
|
|
# Edit (remove the circular ref, keep other deps)
|
|
|
|
|
NEW_BODY=$(echo "$BODY" | sed 's/- #XXX/- None/')
|
fix: Replace Codeberg dependency with local Forgejo instance (#611)
- Add setup_forge() to bin/disinto: provisions Forgejo via Docker,
creates admin + bot users (dev-bot, review-bot), generates API
tokens, creates repo, and pushes code — all automated
- Rename env vars: CODEBERG_TOKEN→FORGE_TOKEN, REVIEW_BOT_TOKEN→
FORGE_REVIEW_TOKEN, CODEBERG_REPO→FORGE_REPO, CODEBERG_API→
FORGE_API, CODEBERG_WEB→FORGE_WEB, CODEBERG_BOT_USERNAMES→
FORGE_BOT_USERNAMES (with backwards-compat fallbacks)
- Rename API helpers: codeberg_api()→forge_api(), codeberg_api_all()
→forge_api_all() (with compat aliases)
- Add forge_url field to project TOML; load-project.sh derives
FORGE_API/FORGE_WEB from forge_url + repo
- Update parse_repo_slug() to accept any host URL, not just codeberg
- Forgejo data stored under ~/.disinto/forgejo/ (not in factory repo)
- Update all 58 files: agent scripts, formulas, docs, site HTML
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-23 16:57:12 +00:00
|
|
|
forge_api PATCH "/issues/NNN" -d "$(jq -nc --arg b "$NEW_BODY" '{body:$b}')"
|
2026-03-16 21:06:50 +01:00
|
|
|
```
|
|
|
|
|
|
|
|
|
|
### Stale dependencies (P3)
|
|
|
|
|
When you see "Stale dependency: #A blocked by #B (open N days)", the dep may be
|
|
|
|
|
obsolete or misprioritized. Investigate:
|
|
|
|
|
|
|
|
|
|
1. Check if dep #B is still relevant (read its body, check if the code it targets changed)
|
|
|
|
|
2. If the dep is obsolete → remove it from #A's `## Dependencies` section
|
|
|
|
|
3. If the dep is still needed → escalate, suggesting to prioritize #B or split #A
|
|
|
|
|
|
|
|
|
|
### Dev-agent blocked (P2)
|
|
|
|
|
When you see "Dev-agent blocked: last N polls all report 'no ready issues'":
|
|
|
|
|
|
|
|
|
|
1. Check if circular deps exist (they'll appear as separate P3 alerts)
|
|
|
|
|
2. Check if all backlog issues depend on a single unmerged issue — if so, escalate
|
|
|
|
|
to prioritize that blocker
|
|
|
|
|
3. If no clear blocker, escalate with the list of blocked issues and their deps
|
|
|
|
|
|
2026-03-12 13:04:50 +00:00
|
|
|
## Escalation
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
|
2026-03-14 16:25:33 +01:00
|
|
|
If you can't fix it, escalate via Matrix:
|
2026-03-12 13:04:50 +00:00
|
|
|
```bash
|
2026-03-14 16:25:33 +01:00
|
|
|
source ${FACTORY_ROOT}/lib/env.sh
|
|
|
|
|
matrix_send "supervisor" "🏭 ESCALATE: <what's wrong and why you can't fix it>"
|
2026-03-12 13:04:50 +00:00
|
|
|
```
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
|
2026-03-12 13:04:50 +00:00
|
|
|
Do NOT escalate if you can fix it. Do NOT ask permission. Fix first, report after.
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
|
2026-03-12 13:04:50 +00:00
|
|
|
## Output
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
|
|
|
|
|
```
|
|
|
|
|
FIXED: <what you did>
|
|
|
|
|
```
|
2026-03-12 13:04:50 +00:00
|
|
|
or
|
|
|
|
|
```
|
|
|
|
|
ESCALATE: <what's wrong>
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
```
|
2026-03-12 13:04:50 +00:00
|
|
|
|
|
|
|
|
## Learning
|
|
|
|
|
|
|
|
|
|
If you discover something new, append it to the relevant best-practices file:
|
|
|
|
|
```bash
|
2026-03-15 18:06:25 +01:00
|
|
|
bash ${FACTORY_ROOT}/supervisor/update-prompt.sh "best-practices/<file>.md" "### Lesson title
|
2026-03-12 13:04:50 +00:00
|
|
|
Description of what you learned."
|
feat: factory supervisor with priorities, auto-fix, and claude -p escalation
- P0: memory crisis (auto-kill stale claude, drop caches, restart Anvil)
- P1: disk pressure (docker prune, log truncate, worktree cleanup, WP log trim)
- P2: factory stopped (CI stuck, dev-agent dead, git broken — auto-fix where possible)
- P3: factory degraded (derailed PRs, auto-trigger reviews)
- P4: housekeeping (stale processes, log rotation)
Calls claude -p only for P0/P1 issues that auto-fix couldn't resolve.
PROMPT.md contains distilled operational knowledge + self-update mechanism.
2026-03-12 13:00:17 +00:00
|
|
|
```
|