Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add liza-mas/liza --skill spec-backfillgit clone --depth 1 https://github.com/liza-mas/lizaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/liza-mas/liza/spec-backfill)<a href="https://agentmods.dev/skills/liza-mas/liza/spec-backfill"><img src="https://agentmods.dev/badge/skills/liza-mas/liza/spec-backfill/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/liza-mas/liza/spec-backfill"><img src="https://agentmods.dev/badge/skills/liza-mas/liza/spec-backfill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 52 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00018 | $0.02069 |
| Opus 5 | $0.00009 | $0.01035 |
| Sonnet 5 | $0.00004 | $0.00414 |
| Haiku 4.5 | $0.00002 | $0.00207 |
Grade A, and why
spec-backfill scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Objective
Reconstruct specifications from a repository's codebase, existing documentation, and git history. You're doing archaeology — finding the features buried in code and docs, then surfacing them as structured specs.
A spec is warranted when code implements user-facing functionality or business logic. Not every module needs a spec. Your job is to find the functional boundaries and document them.
Maintain state in files so work isn't lost if the conversation ends.
Entry Point
- Check for existing session state (
specs/spec-backfill-state.yaml) — if present, offer to resume from cursor position (seereferences/session-state-schema.md) - Check for existing mapping (
specs/spec-mapping.yaml) — determines phase:- No mapping → First Run
- Mapping exists → Incremental
- User specifies files → Targeted
Phases
Three modes of operation. The agent runs the phase that matches the situation.
| Phase | When | Steps |
|---|---|---|
| First run | No spec-mapping.yaml exists |
Classify → Map → Gaps → Generate → Verify |
| Incremental | Mapping exists, detect drift | Detect stale → Reconcile → Generate if needed |
| Targeted | User picks specific files/modules | Classify targets → Map → Generate → Verify |
Phase: First Run
1. CLASSIFY CODE
- Scan codebase, classify each path:
functional (needs spec) | architectural (needs ADR) | utility (needs neither)
- Judgment call: "If a PM asked 'what does this do for users?',
would there be a meaningful answer?" If yes → functional.
- Show classification to user for validation
2. MAP CODE TO SPECS
- For each functional path: does a spec exist? Check alignment.
- No spec → gap. Spec conflicts with code → conflict.
- Conflict types: code-ahead | spec-ahead | semantic-mismatch | stale-spec
- Output: coverage report per references/report-format.md
3. ASK — surface unknowns, prioritize gaps
- "I found N code paths without specs. Which are highest priority?"
- Never generate without confirmation.
4. GENERATE SPECS (per gap, one at a time)
- Extract observable behavior from code
- Identify inputs, outputs, side effects, business rules
- Ask user for: purpose, user story, acceptance criteria
- Generate spec only after confirmation
- Templates: references/spec-templates.md
5. VERIFY
- All functional code has spec mapping
- No relevant legacy content dropped (diff old vs new, surface differences)
- No internal contradictions
- Cross-references valid (ADRs, code paths)
- Archive superseded specs to _archive after confirmation
6. PERSIST
- Update spec-mapping.yaml (including SHAs) per references/mapping-schema.md
- Update changelog
- Output report per references/report-format.md
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 230 lines · 18 tokens per session scan A 6d59e4c48895
spec-backfill is a skill published in the GitHub repository liza-mas/liza (382 stars, last pushed yesterday), licensed Apache-2.0. It adds 18 tokens to every session and 2,069 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
rework-rate
Measure and interpret PR rework rate — the emerging 5th DORA metric.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
full-repo-review
Comprehensive four-wave review of all repo source files, producing a prioritized issue backlog.