Install
$ agentstack add skill-hlhr202-swe-skills-architect-implement ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo 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.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
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 →About
Architect Implement
Use this skill to implement a track that already exists in an initialized Architect project. It executes the selected track's plan.md while keeping architect/tracks.md, track metadata, and project context synchronized.
Core Rules
- Follow the bundled resource
references/implement-track-protocol.mdas the source of truth. - Use Architect semantics consistently:
architect/,/architect-*,Architect methodology, andarchitect(...)commit scopes. - Require an initialized Architect context and a valid track before implementation.
- Use relative paths rooted in the user's project when editing Architect files. Architect-managed files must stay under
architect/; do not use absolute paths,.., or track links outsidearchitect/tracks/. - Ask the user through the active agent runtime's user-interaction mechanism for decisions and workflow confirmations. If structured choices are unavailable, present the options in text and wait for the user's reply. Manual Mode preserves all human confirmations; Auto Mode bypasses phase-level human confirmations after the user selects it.
- Present detailed Markdown, drafts, diffs, reports, verification summaries, or risk analysis in a normal assistant message before asking for a decision. Use interactive prompts only for concise plain-text questions and short plain-text choices.
- Use the active agent runtime's safest reviewable file-editing mechanism, preferably patch-based, for manual file creation and edits. Do not use shell redirection to write files.
- Validate each operation result before continuing. If a step fails, correct it once when the error is clear; otherwise stop and report the blocker.
- Do not commit unless the user explicitly asks for a commit in the current conversation, has explicitly authorized commits for the current implementation workflow, or selected Auto Mode for phase checkpoint commits.
- Do not archive or delete track folders unless the user explicitly confirms that cleanup action.
Execution
- Read the bundled resource
references/implement-track-protocol.md. - Verify required Architect context files exist.
- Parse
architect/tracks.mdand select the target track. - Load the track's
spec.md,plan.md,metadata.json, andarchitect/workflow.md. - Ask whether to run Manual Mode or Auto Mode.
- Execute plan tasks according to the workflow.
- Update task, track, registry, and metadata status.
- Synchronize project documentation when the track is complete.
- Offer review/archive/delete/skip cleanup choices safely.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: hlhr202
- Source: hlhr202/swe-skills
- License: Apache-2.0
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.