Skip to main content
skill CX Codex

eval-speckit-extension — Claude Code skill: how to install

eval-speckit-extension is a Claude Code skill that tests speckit-extension against a completed spec, verifying lifecycle hooks and .spec-context.json timestamps.

MIT ★ 110 updated 2026-10-19

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add alfredoperez/speckit-companion --skill eval-speckit-extension -a claude-code
Open on GitHub

What eval-speckit-extension does

The skill acts as an integration regression suite for the speckit-extension tooling. It inspects a target directory inside specs/ by matching recent conversation commands or picking the folder with the latest .spec-context.json. The agent checks whether each step in the specification workflow produced a valid capture entry.

Verification relies on two Python scripts paired with a conversational audit. First, check_capture.py inspects history[] arrays, monotonic timestamps, and task markers. Then check_quality.py measures text verbosity against size limits and detects abnormal bursts or stalls across steps.

Running the evaluation requires python3 installed on the host machine. The run concludes with a structured verdict table grading seven core lifecycle assumptions with PASS, PARTIAL, or FAIL markers.

What it helps with

  • Validate that lifecycle hooks recorded every workflow step
  • Detect backfilled timestamps and timing outliers in .spec-context.json
  • Audit spec artifact verbosity using check_quality.py
  • Catch duplicated per-task execution logs after implementation

Example prompts for the agent

  • /eval-speckit-extension
  • Verify that lifecycle hooks fired for the current spec
  • Check whether task substeps were captured in .spec-context.json

How to install eval-speckit-extension

1Skills CLI (Vercel)

npx skills add alfredoperez/speckit-companion --skill eval-speckit-extension -a claude-code

2Manually: put the files in place

~/.claude/skills/eval-speckit-extension/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/eval-speckit-extension/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about eval-speckit-extension

How to install eval-speckit-extension in Claude Code?

Run npx skills add alfredoperez/speckit-companion --skill eval-speckit-extension -a claude-code in your terminal. The configuration installs to ~/.claude/skills/eval-speckit-extension/SKILL.md.

What does eval-speckit-extension do?

It evaluates speckit-extension captures against a spec folder. The skill checks lifecycle hooks, canonical history structures, and timing data across the session.

Which scripts does the skill execute?

It runs python3 apps/speckit-extension/scripts/check_capture.py followed by check_quality.py on the target specification directory.

Source text

SKILL.md · MIT
---
name: eval-speckit-extension
description: Evaluate the speckit-extension end-to-end against a real spec — verify lifecycle hooks fired, .spec-context.json captured each step into canonical history[], timing is real (not backfilled), and per-task execution was journaled. Umbrella eval for the spec-kit-side extension; grows as features land (auto-mode, etc.). Use when the user says "/eval-speckit-extension", "verify the extension worked", "did the hooks fire", "check the capture", or just ran a /speckit.* spec and wants it validated.
---

# Eval: speckit-extension

The regression net for the `speckit-extension` (the spec-kit-side half of SpecKit Companion). It verifies the assumptions behind each feature against a **real spec the user created**, combining a deterministic file checker with a conversational cross-check of what the user actually did this session. Re-run it whenever a lifecycle/auto-mode/future feature ships.

Eval areas (add a section per feature as the extension grows):
1. **Lifecycle capture** (shipped) — hooks → `write-context.py` → canonical `history[]`; per-task journaling; derive-from-files fallback.
2. **Command quality** (shipped) — `check_quality.py`: verbosity budgets on the spec artifacts, time-waste checks on `history[]` (untrusted spans, bursts, outliers), static prompting checks on the command-body sources.
3. _Auto-mode_ (future) — add checks here when it lands.
4. _…_

## Steps

### 1. Resolve the target spec

- If the user named a spec (`NNN` or a dir), use `specs/<that>/`.
- Else pick the spec the user most recently created/ran in **this conversation** (whose `/speckit.*` or `/sdd:*` commands appear above). Fall back to the most recently modified `specs/*/` containing `.spec-context.json`.

### 2. Run the deterministic checker

```bash
python3 apps/speckit-extension/scripts/check_capture.py specs/<NNN>-<slug>/
```

Read the PASS/FAIL/INFO rows and the timing breakdown. (`--json` for machine output; `--strict` to exit non-zero on any FAIL.)

### 2b. Run the command-quality checker

```bash
python3 apps/speckit-extension/scripts/check_quality.py --feature-dir specs/<NNN>-<slug> --commands-dir apps/speckit-extension/commands
```

Same report shape plus a **WARN** tier (judgment calls — budgets, duration outliers, untrusted spans; never affects `--strict`, which fails only on FAIL). Three dimensions: **verbosity** (spec/plan/tasks against line/char WARN/FAIL bands calibrated on the completed specs 484/509/510 — only oversize flags; a missing or lean artifact is fine), **time-waste** (each reached step's span trusted only with ordered extension-stamped boundaries; ≥3 `by:ai` task finishes inside one second FAIL as the pre-#509 burst shape; a step >8× the median of the others and >5m WARNs), and **prompting** (never-halts commands — the after-* hooks, living-drift/sync/coverage, mark-complete, status, resume, classify — must contain no ask-the-user instruction; the clarify carrier must ask; negated mentions and fenced templates don't count; a missing roster file fails loudly). CI runs all three strict over `specs/509-timing-capture`, `specs/510-living-sync`, and the shipped command sources.

### 3. Conversational cross-check

The checker can't know what the user *did*. Reconstruct it from this conversation and reconcile against the capture:

- List every `/speckit.specify|plan|tasks|implement` (or `/sdd:*`) the user ran this session, with rough timestamps.
- For each, confirm a matching `history[]` entry exists with the expected `step`/`status` and `by: "extension"` (hook fired) — or `by: "ai"` if the AI appended it, `by: "derive"` if reconstructed. (The canonical `by` vocabulary is `extension`/`user`/`cli`/`ai`/`derive`.)
- **Timing reality:** does each capture's `at` land *just after* the command ran (live hook), or are timestamps clustered/round (backfilled)? The `timestamps-real` check flags round-ms; you add the "fired at the right moment" judgement.
- **Task execution:** if `/speckit.implement` ran, confirm per-task substep entries appear. Read `task-cadence`'s **source**: `live (by:ai)` means the AI journaled each task as it finished — non-zero gaps are the real-cadence signal; `hook burst (by:extension)` means the single end-of-step hook synced them — 0ms gaps are *expected* there and not a defect (the AI didn't journal live, so the hook backstopped). Only a *mixed* or duplicated picture is a smell.

### 4. Score the assumptions

Render a verdict table (PASS / PARTIAL / FAIL + one-line evidence):

| # | Assumption (lifecycle capture) |
|---|------------|
| A1 | Each lifecycle step the user ran produced a capture (`after_specify/plan/tasks/implement` → a `history[]` entry). |
| A2 | Writes use canonical `history[]` with explicit `kind` — no legacy `transitions[]`/`stepHistory`. |
| A3 | Timing is real **for deterministic writes** (`by:extension`/`derive`/`cli`/`user`): ms-precision and monotonic (`timestamps-real`/`timestamps-monotonic` check these only). `by:ai` entries carry second precision (`date -u +%SZ`) and may burst — that's graded by `task-cadence`, not failed. |
| A4 | `/speckit.implement` journaled per-task progress as implement **substeps** (`substep == task id`), matching `tasks.md` completed markers. |
| A5 | No-backward-clobber held — no step regressed; an advanced/terminal spec was never dragged back. |
| A6 | (On demand) `derive-from-files.py` reconstructs the same state from artifacts when a hook didn't fire. Test: back up `.spec-context.json`, delete it, run `python3 apps/speckit-extension/scripts/derive-from-files.py --feature-dir specs/<NNN>-<slug>`, diff, restore. |
| A7 | Per-task journaling is **deduped** — each task carries one `start` + one `complete`, so dedup is checked per `(task, kind)` (`per-task-no-duplicates`); a repeated `(task, kind)` means the `after_implement` hook re-added an already-journaled entry. Live AI entries (`by: ai`, real `date -u` timing) are the cadence source. Real per-task cadence requires `task-cadence` source `live (by:ai)` with non-zero gaps; a `hook burst` source means the AI didn't journal live and the hook backstopped (correct final state, coarse timing — not a defect). |

End with a one-paragraph plain verdict: did the extension work, and where reality diverged from the assumptions.

## Extending

When a feature ships:
- add a deterministic assertion to `check_capture.py` (`run_checks`) and, if it's a new area, a new eval section + assumption block above,
- keep `VALID_BY` / `CANONICAL_STEPS` / `CANONICAL_STATUSES` in `check_capture.py` in sync with `apps/vscode/src/core/types/spec-context.schema.json` (the jest drift guard locks schema↔TS; this is the eval side).

© alfredoperez · MIT · alfredoperez/speckit-companion

Similar skills

All skills for Testing and QA →

From the same set83

The whole set →
publish-canvas publish-canvas ships the Copilot app canvas: version bump, tests, the canvas-v<X.Y.Z> tag, then the github/awesome-copilot listing update. CX Codex +1 ★ 110repo stars bench-replay The bench-replay command for Claude Code and Codex replays one pipeline step from a saved stage and grades the effect of a change. CX Codex +1 ★ 110repo stars speckit.tasks The speckit.tasks prompt turns feature design documents into a dependency-ordered tasks.md, grouped by user story. Use it to prepare an implementation plan from an existing spec. CX Codex GP ChatGPT ★ 110repo stars changelog-entry changelog-entry is a Claude Code command that writes a changelog entry from a PR or current diff and chooses its file and format. Run it with a PR number or no argument. CX Codex +1 ★ 110repo stars bench-experiment bench-experiment is a Claude Code command that runs a Conduit benchmark round and compares results with earlier sweeps to measure changes across selected cells. CX Codex +1 ★ 110repo stars speckit.analyze speckit.analyze runs a read-only pass over spec.md, plan.md and tasks.md, then prints a report on contradictions, duplicates and requirements that no task covers. CX Codex GP ChatGPT ★ 110repo stars speckit.companion.living-sync speckit.companion.living-sync updates living specs to match current working-tree changes in a single pass. Modifies existing spec files without regenerating them from scratch. CX Codex +1 ★ 110repo stars speckit.implement The speckit.implement prompt turns the plan in tasks.md into an ordered implementation run. It is listed for Claude Code and Codex. CX Codex GP ChatGPT ★ 110repo stars

Project sponsors

Companies that keep this analytics open