openspec-verify-change
Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
- License 许可
- MIT
- 所属仓库
- psacc/omnisess
- 最近更新
- 2026-11-12
- Stars
- ★ 1
- Forks
- 1
01 前言 Front Matter
- name
- openspec-verify-change
- license
- MIT
- metadata
- { "author": "openspec", "version": "1.0", "generatedBy": "1.1.1" }
- description
- Verify implementation matches change artifacts. Use when the user wants to validate that implementation is complete, correct, and coherent before archiving.
- compatibility
- Requires openspec CLI.
02 资源文件 5
- 脚本Tools/FetchFredSeries.ts12.1 KB
- 脚本Tools/GenerateAnalysis.ts15.8 KB
- 脚本Tools/UpdateSubstrateMetrics.ts22.9 KB
- 参考文档Workflows/GetCurrentState.md7.8 KB
- 参考文档Workflows/UpdateData.md3.6 KB
03 SKILL.md 正文
Verify that an implementation matches the change artifacts (specs, tasks, design).
Input: Optionally specify a change name. If omitted, check if it can be inferred from conversation context. If vague or ambiguous you MUST prompt for available changes.
Steps
-
If no change name provided, prompt for selection
Run
openspec list --jsonto get available changes. Use the AskUserQuestion tool to let the user select.Show changes that have implementation tasks (tasks artifact exists). Include the schema used for each change if available. Mark changes with incomplete tasks as "(In Progress)".
IMPORTANT: Do NOT guess or auto-select a change. Always let the user choose.
-
Check status to understand the schema
openspec status --change "<name>" --jsonParse the JSON to understand:
schemaName: The workflow being used (e.g., "spec-driven")- Which artifacts exist for this change
-
Get the change directory and load artifacts
openspec instructions apply --change "<name>" --jsonThis returns the change directory and context files. Read all available artifacts from
contextFiles. -
Initialize verification report structure
Create a report structure with three dimensions:
- Completeness: Track tasks and spec coverage
- Correctness: Track requirement implementation and scenario coverage
- Coherence: Track design adherence and pattern consistency
Each dimension can have CRITICAL, WARNING, or SUGGESTION issues.
-
Verify Completeness
Task Completion:
- If tasks.md exists in contextFiles, read it
- Parse checkboxes:
- [ ](incomplete) vs- [x](complete) - Count complete vs total tasks
- If incomplete tasks exist:
- Add CRITICAL issue for each incomplete task
- Recommendation: "Complete task: <description>" or "Mark as done if already implemented"
Spec Coverage:
- If delta specs exist in
openspec/changes/<name>/specs/:- Extract all requirements (marked with "### Requirement:")
- For each requirement:
- Search codebase for keywords related to the requirement
- Assess if implementation likely exists
- If requirements appear unimplemented:
- Add CRITICAL issue: "Requirement not found: <requirement name>"
- Recommendation: "Implement requirement X: <description>"
-
Verify Correctness
Requirement Implementation Mapping:
- For each requirement from delta specs:
- Search codebase for implementation evidence
- If found, note file paths and line ranges
- Assess if implementation matches requirement intent
- If divergence detected:
- Add WARNING: "Implementation may diverge from spec: <details>"
- Recommendation: "Review <file>:<lines> against requirement X"
Scenario Coverage:
- For each scenario in delta specs (marked with "#### Scenario:"):
- Check if conditions are handled in code
- Check if tests exist covering the scenario
- If scenario appears uncovered:
- Add WARNING: "Scenario not covered: <scenario name>"
- Recommendation: "Add test or implementation for scenario: <description>"
- For each requirement from delta specs:
-
Verify Coherence
Design Adherence:
- If design.md exists in contextFiles:
- Extract key decisions (look for sections like "Decision:", "Approach:", "Architecture:")
- Verify implementation follows those decisions
- If contradiction detected:
- Add WARNING: "Design decision not followed: <decision>"
- Recommendation: "Update implementation or revise design.md to match reality"
- If no design.md: Skip design adherence check, note "No design.md to verify against"
Code Pattern Consistency:
- Review new code for consistency with project patterns
- Check file naming, directory structure, coding style
- If significant deviations found:
- Add SUGGESTION: "Code pattern deviation: <details>"
- Recommendation: "Consider following project pattern: <example>"
- If design.md exists in contextFiles:
-
Generate Verification Report
Summary Scorecard:
## Verification Report: <change-name> ### Summary | Dimension | Status | |--------------|------------------| | Completeness | X/Y tasks, N reqs| | Correctness | M/N reqs covered | | Coherence | Followed/Issues |Issues by Priority:
-
CRITICAL (Must fix before archive):
- Incomplete tasks
- Missing requirement implementations
- Each with specific, actionable recommendation
-
WARNING (Should fix):
- Spec/design divergences
- Missing scenario coverage
- Each with specific recommendation
-
SUGGESTION (Nice to fix):
- Pattern inconsistencies
- Minor improvements
- Each with specific recommendation
Final Assessment:
- If CRITICAL issues: "X critical issue(s) found. Fix before archiving."
- If only warnings: "No critical issues. Y warning(s) to consider. Ready for archive (with noted improvements)."
- If all clear: "All checks passed. Ready for archive."
-
Verification Heuristics
- Completeness: Focus on objective checklist items (checkboxes, requirements list)
- Correctness: Use keyword search, file path analysis, reasonable inference - don't require perfect certainty
- Coherence: Look for glaring inconsistencies, don't nitpick style
- False Positives: When uncertain, prefer SUGGESTION over WARNING, WARNING over CRITICAL
- Actionability: Every issue must have a specific recommendation with file/line references where applicable
Graceful Degradation
- If only tasks.md exists: verify task completion only, skip spec/design checks
- If tasks + specs exist: verify completeness and correctness, skip design
- If full artifacts: verify all three dimensions
- Always note which checks were skipped and why
Output Format
Use clear markdown with:
- Table for summary scorecard
- Grouped lists for issues (CRITICAL/WARNING/SUGGESTION)
- Code references in format:
file.ts:123 - Specific, actionable recommendations
- No vague suggestions like "consider reviewing"
04 同仓库更多 Skills
stats
Show tool-call counts and file-I/O activity from the transcript SQLite index. Aggregates Read/Write/Edit file paths, line counts, token usage, and error flags across a time window (default 7d) or for a single session. Use this skill when asked which tools were used most, which files saw the most activity, how many tokens were spent, or to inspect a single session's tool-call summary. Trigger phrases: "session stats", "tool stats", "what tools did I use", "tool usage", "files touched", "token usage", "stats for last week", "summarize tool calls", "session tool breakdown".
skills-audit
Classify agent skills by usage tier (Keep / Borderline / Archive / Unknown) using Claude Code session JSONL transcripts as the signal source. Read-only — never modifies skills, sessions, or any file on disk. Use this skill when asked to audit skills, find unused skills, classify skills, check which skills get used, produce a skill usage report, or prune skills. Trigger phrases: "audit skills", "find unused skills", "classify skills", "which skills do I use", "skill usage report", "prune skills", "what skills are unused", "skills I can remove", "trim my skill list". Output is a recommendation (Keep / Borderline / Archive) — acting on it is always a manual step.
show
Show the full detail of a single AI coding session by its ID. Session IDs are in the format <tool>:<id> (e.g., claude:5c3f2742, cursor:a1b2c3d4). Use this skill when asked to show a specific session, inspect a session, read a session, or view the messages in a session. Trigger phrases: "show session", "open session", "inspect session", "read session", "show me claude:5c3f2742", "view session details", "what happened in session".
search
Full-text search across all AI coding sessions (Claude Code, Cursor, Codex, GitHub Copilot CLI). Use this skill when asked to search sessions, find a conversation about a topic, look up what was discussed in past sessions, or find sessions mentioning specific code or concepts. Trigger phrases: "search sessions", "find sessions about", "look up in sessions", "what did I discuss about", "find conversation about", "search my history".
recap
Produce a structured markdown briefing of AI coding sessions for a given time window. Groups sessions by project and summarizes each group with intent, outcome, and status. Use this skill when asked to recap recent coding work, summarize today's sessions, show what was worked on, or produce a daily/weekly session summary. Trigger phrases: "recap", "what did I work on", "daily summary", "session summary", "summarize my sessions", "what sessions did I have today", "coding recap". This is the sessions-only public version — no Slack, Jira, or Granola integration.
ps
Show live Claude Code and Codex sessions as a process tree, grouped by shared ancestor chain up to launchd. Covers Claude CLI sessions, Claude Desktop agent mode, and Codex sessions (CLI TUI + Codex.app threads). Use this skill when asked about the process tree, ancestor lineage, which terminal or tmux window a Claude or Codex session is running in, or what's currently running. Trigger phrases: "process tree", "what's running", "lineage", "ancestor chain", "which terminal is my claude in", "tmux tree", "claude process tree", "codex process tree", "session lineage". macOS-only: on other platforms the command prints "unsupported" and exits 0.