signal-discovery — Claude Code skill: signal audits
The signal-discovery Claude Code skill finds gaps in AI watermark and provenance detection, then checks removal parity. It runs audits of fresh or retained corpora for the remove-ai-watermarks library.
Apache-2.0 ★ 6,481 updated 2026-10-24
npx skills add wiltodelta/remove-ai-watermarks --skill signal-discovery -a claude-codeWhat signal-discovery does
The skill audits which AI marks and formats remove-ai-watermarks can detect, then sorts missed cases into actionable gaps. It compares recent data with the retained history and checks whether newly detected removable signals are cleared without breaking fidelity.
Run it from the repository root. The workflow starts with the library’s development rules, verification plan, and supported-signal documentation. Fresh data requires the external collection process that owns the corpus. Retained inputs stay read-only; reports and derived files belong in the Git-ignored .local-eval/ directory and must not be committed.
What it helps with
- find new AI mark providers and formats in recent data
- rerun detection across retained history after a code change
- check that detected marks are removed without violating fidelity
- triage parsing errors and missing provider attribution
Example prompts for the agent
- Audit recent signals in
remove-ai-watermarksand sort the coverage gaps. - Rerun discovery across the full retained history after the detector changes.
- Check detection and removal parity for the newly detected signals.
How to install signal-discovery
1Skills CLI (Vercel)
npx skills add wiltodelta/remove-ai-watermarks --skill signal-discovery -a claude-code2Manually: put the files in place
~/.claude/skills/signal-discovery/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/signal-discovery/
1Skills CLI (Vercel)
npx skills add wiltodelta/remove-ai-watermarks --skill signal-discovery -a codex2Manually: put the files in place
.agents/skills/signal-discovery/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add wiltodelta/remove-ai-watermarks --skill signal-discovery -a cursor2Manually: put the files in place
.cursor/skills/signal-discovery/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install wiltodelta/remove-ai-watermarks .claude/skills/signal-discovery2Manually: put the files in place
.github/skills/signal-discovery/SKILL.md1Skills CLI (Vercel)
npx skills add wiltodelta/remove-ai-watermarks --skill signal-discovery -a gemini-cli2Platform CLI
gemini skills install https://github.com/wiltodelta/remove-ai-watermarks --path .claude/skills/signal-discoveryManually: put the files in place
.devin/skills/signal-discovery/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add wiltodelta/remove-ai-watermarks --skill signal-discovery -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about signal-discovery
How do I install signal-discovery in Claude Code?
Run npx skills add wiltodelta/remove-ai-watermarks --skill signal-discovery -a claude-code. The skill file is at ~/.claude/skills/signal-discovery/SKILL.md.
What does signal-discovery do?
It looks for missing AI provenance and watermark coverage in remove-ai-watermarks, then tests detection and removal against fresh or historical data.
What does signal-discovery need to run?
Run it from the remove-ai-watermarks repository root. Refreshing the corpus requires its external collection process, and retained inputs must remain read-only.
Source text
SKILL.md · Apache-2.0
--- name: signal-discovery description: Discover missing AI provenance and watermark coverage in the remove-ai-watermarks library from fresh or historical corpora, then validate detection and removal parity. Use for corpus gap mining, periodic signal audits, new-provider discovery, or deciding whether old retained data must be replayed. Do not use for ordinary per-file identification or removal. metadata: internal: true --- # Signal discovery Work from the `remove-ai-watermarks` repository root. This skill discovers and closes library coverage gaps. Implementation, tests, fixtures, and documentation changes belong in this library; do not patch a downstream consumer to compensate for a detector or cleaner gap. Treat retained inputs as sensitive and read-only. Keep reports, contact sheets, sidecars, and derived samples under the library's gitignored `.local-eval/`. Never commit a retained input or an artifact derived from one. Before acting, read: - `.claude/rules/development.md`; - `docs/verification-plan.md`; - `docs/supported-signals.md` and `docs/watermarking-landscape.md`; - [references/coverage-and-cadence.md](references/coverage-and-cadence.md). ## Choose the run Do not call `corpus_gap_scan.py` a complete signal audit. Select every mode the question requires: 1. **Fresh discovery** finds new providers, serializations, labels, and formats in objects added since the preceding review. 2. **Historical replay** reruns current code on the complete retained history. Use it after detection-path or format-support changes and on the periodic backstop defined in the reference. 3. **Removal parity** proves that every detected removable signal is cleared without violating the signal family's fidelity invariant. 4. **Calibration or provider oracle** is required where the corpus has no self-evident answer key. Never treat a local negative as proof that a proprietary watermark is absent. Start by inventorying the current code, not a remembered list. Enumerate image and video visible-mark registries, metadata readers and strippers, local invisible decoders, explicit invisible vendor routes, supported containers, and the stable signal families emitted by `identify`. Reconcile that inventory with the coverage matrix before interpreting any green result. A hand-written marker list or report family map that omits a current registry member is itself a finding. ## Run protocol 1. Record the library commit, version, enabled extras, corpus date range, and exact commands in an untracked run directory. Preserve the pre-run Git status so unrelated work is not attributed to the audit. 2. Refresh the local retained corpus through its owning, external collection process when the task calls for fresh data. Use an explicit gitignored destination, do not delete local objects, and finish with a zero-delta pass. Fetch paired outputs only for a removal or complaint investigation. 3. Run the applicable fresh and historical discovery modes from the reference. Use `corpus_gap_scan.py --since YYYY-MM-DD` for the overlapping fresh window and omit `--since` for the complete history. Report counts per input shape and per outcome; an example does not establish coverage. Give every code and corpus snapshot a distinct `--report` path. The adjacent `.progress.jsonl` checkpoint is the live progress source and makes the same command resumable; use `--restart` only when intentionally discarding it. 4. Triage every candidate into one of: parser/serialization gap, missing vendor attribution, new visible mark, possible invisible signal, format/read error, expected non-AI provenance, or false candidate. Preserve unresolved items as unresolved. 5. Reproduce an actionable gap with a failing test before changing code. Derive the smallest publication-safe fixture from a controlled source or construct it synthetically; do not move the retained file into a tracked path. 6. Implement the fix in the library subsystem that owns it. Update the library registry, parser, remover, public docs, and published agent skill when their contracts move. Updating a downstream consumer to take a released version is a later, separately scoped task. 7. Rerun the affected focused audit and the complete historical regression mode. A detection change must name the files expected to move before the after-run; classify lost signals separately from intended new detections. 8. Run removal parity for every newly detected removable signal. A quiet post-removal detector is necessary but not always sufficient: check pixel/stream invariants and visually inspect residual-prone visible marks. 9. Run `bash maintain.sh`. Report evidence limits, incomplete rows, crashes, unsupported formats, and oracle-unavailable cohorts explicitly. When a cohort needs a provider verdict, walk the `provider-oracles` skill's plan automatically, by its order and routes, without asking before switching.
© wiltodelta · Apache-2.0 · wiltodelta/remove-ai-watermarks