Quality Gates (Self-Review Before Seal)

Quality Gates (Self-Review Before Seal)

After writing findings, perform ONE revision pass:

1. Re-read your output file
2. For each P1 finding:
   • Is the evidence actual (code snippet, file:line reference)?
   • Is the issue realistic (not purely theoretical)?
   • Does the file:line reference exist?
3. Weak evidence → re-read source → revise, downgrade, or delete
4. Self-calibration: if you found 0 issues in a high-risk area, broaden your lens

This is ONE pass. Do not iterate further.

Confidence Calibration

• PROVEN: You Read() the file, traced the logic, and confirmed the behavior
• LIKELY: You Read() the file, the pattern matches a known issue, but you didn't trace the full call chain
• UNCERTAIN: You noticed something based on naming, structure, or partial reading — but you're not sure if it's intentional

Rule: If >50% of findings are UNCERTAIN, you're likely over-reporting. Re-read source files and either upgrade to LIKELY or move to Unverified Observations.

Inner Flame (Supplementary)

After the revision pass above, verify grounding:

• Every file:line cited — actually Read() in this session?
• Weakest finding identified and either strengthened or removed?
• All findings valuable (not padding)?

Include in Self-Review Log: "Inner Flame: grounding={pass/fail}, weakest={finding_id}, value={pass/fail}"