AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified Apache-2.0 Self-run

Feature Loop

skill-demondamon-agenticx-feature-loop · by DemonDamon

Coding agent for the project-level harness — one feature per session, must verify, must be mergeable.

No reviews yet
0 installs
28 views
0.0% view→install

Install

$ agentstack add skill-demondamon-agenticx-feature-loop

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-demondamon-agenticx-feature-loop)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Feature Loop? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Feature Loop

Use when session_mode == feature_loop and the project is past the Initializer phase (system prompt will say "Coding 阶段").

Iron rules

  1. One feature per session. Multiple in_progress features are forbidden by the store; do not bypass.
  2. Always verify before commit. feature_complete rejects features that have not been promoted to verified via verify_run.
  3. Keep the branch mergeable. Each commit must compile, pass verify.yaml, and not depend on uncommitted local state.

Loop

  1. Sync from disk. Call project_status first thing every session — it reads feature_list.json / status.json / progress.md and reseeds your understanding from the single source of truth.
  2. Pick a feature.
  • If status.active_feature_id is already set, resume that one (the prompt block will tell you).
  • Otherwise call feature_select with no arguments to auto-pick the highest-priority pending feature with satisfied dependencies, or pass feature_id when the user names one.
  1. Implement using code_dev phases (Explore → Read → Author):
  • Explore with code_outline, grep, optional code_search.
  • Read with file_read slices (start/end line). Track files you have read in scratchpad.
  • Author with file_write skeletons first, then section-by-section appends.
  1. Run the gate. Call verify_run feature_id=. If any step fails:
  • Do not call feature_complete.
  • Log a progress_append with the failing step name and root cause.
  • Decide: fix and rerun verify_run, or escalate to the user with the specific failure.
  1. Commit via shell.

`` git add -A git commit -m "feat(): " ` Capture the resulting sha (git rev-parse HEAD`).

  1. Promote to committed. Call feature_complete feature_id= commit_sha=. The store writes an immutable archive snapshot under .agx/project/archive/feature_.json.
  2. Decide next step. Either call feature_select for the next feature in the same session, or stop and tell the user the loop closed cleanly.

Failure modes

  • verify_run timeouts: do not retry blindly. Inspect the log under .agx/project/archive/, fix the root cause, then rerun.
  • Dependency missed: feature_select will reject features whose depends_on isn't committed. Either pick a different feature or finish the prerequisite first.
  • Lost context after window compaction: re-call project_status. Disk is the source of truth.
  • Cold start (new machine, fresh session): project_status + feature_select is enough to resume. No memory of previous turns is required.

Forbidden

  • Editing files inside .agx/project/archive/ (immutable).
  • Calling project_init outside the Initializer phase.
  • Skipping verify_run because "tests are too slow".

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.