Contextualizer Self Review

Purpose

Before delivering context files or briefs, the Contextualizer evaluates its own output against the TRACE rubric. Each letter is scored 0, 1, or 2 with evidence quoted from the rubric and cited from actual work. The total determines whether to deliver, rewrite, or abort.

Procedure

1. Gather evidence. Honest self-review makes verification efficient. Accurate scorecards confirm fast; dishonest ones fail and re-run, wasting time and compute. Before scoring, run verification commands to gather proof. The examples below show common patterns — choose what provides the best evidence for your specific work.

   Examples:

   • Schema compliance: verify .context.md files have opening <context> tag with path and date, Summary, Constraints, Guidance sections
   • Feature map: verify FEATURE-MAP.md entries have feature name, flow steps with file paths and role descriptions
   • Anchored claims: test -f <path> for directories/files mentioned in context — confirms claims are grounded
   • Coverage: find . -type f | wc -l and find . -type d | wc -l to check project size against yield threshold
   • Brevity: wc -l <context-files> — each context file should be shorter than the directory it describes

   These are examples, not mandates. Choose commands that provide the strongest proof for your output.

2. Score each criterion. Read the TRACE rubric below. For each letter, assign a score of 0, 1, or 2. You must:

   • Quote the specific criterion level (0, 1, or 2) that your work matches.
   • Cite evidence from your actual work — files verified, schema elements checked, line counts measured. Generic claims like "I verified everything" are not evidence and score 0.

3. Output the Scorecard. Fill in the scorecard below. This is not internal reasoning — this is your deliverable checkpoint.

4. Apply the hard-fail rule. If any letter scores 0, do not deliver — go to step 5 immediately.

5. Determine action by total score:

   • 9 – 10 — DELIVER — Output meets all criteria. Deliver to user.
   • 7 – 8 — FIX the scored < 2 criteria. a. Identify which letters scored below 2. b. Fix those gaps automatically (do NOT consult the user). c. Re-score, then deliver if 9-10. d. If still below 9, retry once more. e. After 2 failed fix attempts, yield with the current state, rubric scores, and blocking letters.
   • 0 – 6 — RESTART — The output is fundamentally broken. Rewrite from scratch with corrected understanding, or yield to the user with an explanation of what went wrong.

Scorecard

Complete this before delivering. Each letter requires the matched criterion quote and specific evidence from your work.

• T — TASK — Score: [0/1/2] — Matched: [quote the 0, 1, or 2 description] — Evidence: [task classification, deliverable type produced]
• R — RUN — Score: [0/1/2] — Matched: [quote the 0, 1, or 2 description] — Evidence: [schema elements verified, skill followed]
• A — ANCHORED — Score: [0/1/2] — Matched: [quote the 0, 1, or 2 description] — Evidence: [files verified with test -f, claims traced to code]
• C — COVERAGE — Score: [0/1/2] — Matched: [quote the 0, 1, or 2 description] — Evidence: [project size checked, yield condition evaluated, incremental vs rewrite]
• E — EVIDENT — Score: [0/1/2] — Matched: [quote the 0, 1, or 2 description] — Evidence: [line counts, brevity comparison]

Total: X/10 → Action: [DELIVER/FIX/RESTART]

TRACE Rubric

T — TASK

Did I classify the task correctly and produce the right deliverable?

• 0 — Wrong mode chosen (full scan when structural brief was needed, or vice versa). Produced the wrong deliverable type entirely.
• 1 — Right mode chosen but one deliverable element is missing or wrong format (e.g., review blocks without LOC counts, structural brief missing Information Flow section).
• 2 — Correct mode, correct deliverable, correct format. Full scan produced .context.md files AND FEATURE-MAP.md. Structural brief has all three sections (Modules, Boundaries, Information Flow). Review blocks respect 1500 LOC limit, module co-location, and boundary rules.

R — RUN

Did I follow the mandated skill and produce output in the correct schema?

• 0 — Did not follow skills/context-maintenance/SKILL.md for full scan. Used ad-hoc format instead of the prescribed .context.md or FEATURE-MAP.md schema. Did not run the directory scan script when producing context files. Structural brief does not use the required three-section format.
• 1 — Followed the skill but one schema detail is off: <context> tag missing path or date, feature flow step lacks "what happens here" description, or updated date touched without content change.
• 2 — Followed skills/context-maintenance/SKILL.md end-to-end. .context.md files use exact schema: opening <context> tag with path and date, Summary, Constraints, Guidance sections. FEATURE-MAP.md uses exact schema: feature name, flow steps in order with file paths and role descriptions. Structural brief uses exact three-section format. Updated only drifted features in the existing map — did not rewrite it.