Install Sepo agent infrastructure

AS
2d ago
+62857-0

Summary

Adds the Sepo agent backend to this repository so @sepo-agent can answer, implement, review, and fix work on GitHub. This PR adds agent infrastructure only — no application source, secrets, or branch protection are changed.

Installed from the latest stable Sepo release v0.5.0 (self-evolving/repo).

Required setup after merge

  1. Install the Sepo GitHub App on this repository, or choose another auth path from the setup guide.
  2. Add at least one model-provider secret under Actions secrets: OPENAI_API_KEY, CLAUDE_CODE_OAUTH_TOKEN, and/or ANTHROPIC_API_KEY.
  3. Run the Agent / Onboarding / Check Setup workflow.
  4. Review the Sepo setup check issue it opens and complete any remaining setup it reports.
  5. If agent/memory is missing, run Agent / Memory / Initialization.
  6. Optionally run Agent / Rubrics / Initialization if this repo wants rubric steering.

Also confirm Issues and GitHub Actions are enabled (Settings → General → Features, Settings → Actions → General).

Source revision

  • self-evolving/repo @ v0.5.0 (5f7d2032080e9c0578d0e4cb3b98b2f695f33547)

Installed files

  • .agent/ — agent backend (265 files), excluding generated .agent/dist/ and .agent/node_modules/.
  • .github/ — agent workflows (28), composite actions, prompts, issue templates, and assets, merged into the existing .github/.
  • .gitignore — appended .agent/dist/ and .agent/node_modules/.

Preserved / skipped

  • All target-owned .github/ files preserved; the conflict audit found no source path that collides with an existing target path.
  • Root README.md, AGENTS.md, and .skills/ were not copied (install defaults).
  • No application source changed.

Validation

  • npm ci, npm run build, and npm test pass on the source at v0.5.0 (run once against the identical payload).
  • git diff --cached --check is clean.
  • Conflict audit: no source .github path collides with an existing target path.

Skipped checks

  • Per-repo .agent build/test not rerun; the payload is byte-identical to the validated source revision.

This PR was created by an AI agent (OpenHands) on behalf of @ashcoft.


Devin Review


Summary by cubic

Installs the Sepo agent backend so @sepo-agent can answer, implement, review, and fix work on this repository. Adds .agent/ and merges .github/ workflows, actions, and prompts from Sepo v0.5.0; no application source, secrets, or branch protection are changed.

Setup after merge

  • Install the Sepo GitHub App on this repository.
  • Add at least one model-provider secret under Actions secrets: OPENAI_API_KEY, CLAUDE_CODE_OAUTH_TOKEN, or ANTHROPIC_API_KEY.
  • Run the Agent / Onboarding / Check Setup workflow and complete anything it reports.
  • Run Agent / Memory / Initialization if agent/memory is missing.
  • Confirm Issues and GitHub Actions are enabled in repository settings.

Generated outputs (.agent/dist/, .agent/node_modules/) are gitignored. Root README.md, AGENTS.md, and .skills/ were not copied.


This change is Reviewable

RetriggerView in GreptileConfidence Score: 1/5

Fix the four workflow issues before merging, especially the checkout failure for existing PRs and the schedule switch that does not stop issue closure.

Fix All in CodexFindings

  1. P1 Existing PR fixes fail ▶
  2. P1 Failed pushes trigger another review ▶
  3. P1 Unsaved scans count as complete ▶
  4. P1 Disabled schedules still close issues ▶
Fix with agent prompt
### Issue 1
.github/workflows/agent-fix-pr.yml:153-159
`checkout-pr.js` replaces the runtime checkout with the PR branch before the workflow loads `./.github/actions/run-agent-task`. Existing PRs branched before this installation lack that action and the new prompts, so `/fix-pr` fails before the agent runs. Keep the runtime on the default branch and check out the PR in a separate worktree, then pass that path through `agent_cwd`.

### Issue 2
.github/workflows/agent-fix-pr.yml:330
The handoff sends `success` even when committing or pushing the fixes fails. Because this step uses `always()`, a rejected push-such as when someone updates the PR during the run-still starts another review in heuristic mode, although the fixes never reached GitHub. Include the outcomes of `commit` and `push-head` in `SOURCE_CONCLUSION`, as the status comment already does.

### Issue 3
.github/workflows/agent-memory-scan.yml:132-133
`last_scan_at` advances even when the scan's memory push fails. `run-agent-task` treats that failure as a warning, so `steps.scan.outcome` remains successful. Concurrent memory writers use separate groups and push without retrying, making a rejected push a realistic case. Later scheduled scans then skip until newer activity appears, although the notes were never saved.

Expose whether saving memory succeeded and require that result before advancing `last_scan_at`. A clean no-change result should still count as success.

### Issue 4
.github/workflows/agent-close-stale-issues.yml:17-22
This daily workflow ignores `AGENT_SCHEDULE_POLICY`. Setting `default_mode` to `disabled`, or disabling `agent-close-stale-issues.yml` specifically, still lets it close issues. The new docs describe that policy as the switch for scheduled runs. Apply the shared scheduled gate before closing issues, while keeping manual dispatch available.

---

For each issue above, determine whether it is valid and should be fixed. If so, fix it directly.

Summary

Installs Sepo's runtime, GitHub workflows, prompts, documentation, and tests without changing application source.

  • GitHub requests can now run through Sepo agent workflows.

Diagram

%%{init: {'theme': 'neutral'}}%%
flowchart TD
  A[GitHub mention or label] --> B[Read request and check access]
  B --> C[Answer]
  B --> D[Implement]
  B --> E[Review]
  B --> F[Fix PR]
  D --> G[Verify and publish]
  F --> G
  G --> H[Optional next job]
  E --> H
  I[Scheduled jobs] --> J[Sync and curate memory]
  I --> K[Maintain issues and check updates]

Reviews (1) · Last reviewed commit: "chore: install Sepo agent infrastructure" · Reviewed by Greptile