Borrowing it
Nothing to install: this file belongs to foundryside-dev/plainweave. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/foundryside-dev/plainweave/main/.agents/skills/wardline-gate/SKILL.mdgit clone --depth 1 https://github.com/foundryside-dev/plainweaveWrote 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/foundryside-dev/plainweave/wardline-gate)<a href="https://agentmods.dev/skills/foundryside-dev/plainweave/wardline-gate"><img src="https://agentmods.dev/badge/skills/foundryside-dev/plainweave/wardline-gate/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/foundryside-dev/plainweave/wardline-gate"><img src="https://agentmods.dev/badge/skills/foundryside-dev/plainweave/wardline-gate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00069 | $0.00881 |
| Opus 5.5 | $0.00028 | $0.00352 |
| Sonnet 5.5 | $0.00014 | $0.00176 |
| Haiku 4.5 | $0.00007 | $0.00088 |
Grade A, and why
wardline-gate 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 1mo 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Wardline: the trust-boundary gate
Wardline is a deterministic, whole-program static taint analyzer. It marks trust
boundaries with two decorators from wardline.decorators: @external_boundary
(untrusted data arriving from outside) and @trusted (a producer that must only
receive validated data). When untrusted data reaches a trusted producer it raises
PY-WL-101 at ERROR.
The loop
- Scan. Run
wardline scan . --fail-on ERROR(or call thescanMCP tool). Read the gate verdict and the active (non-suppressed) findings —activeis the population the gate enforces on. - Explain. For each active defect, call
explain_taint(MCP) or runwardline explain-taint <fingerprint> [PATH](CLI) with the finding'sfingerprint, and itsqualnameassink_qualname. Do this right after the scan and before editing — a stale fingerprint returns an error. With a Loomweave store configured, passchain: true(--chainon the CLI) to walk the full taint chain back to the originating boundary. - Fix at the BOUNDARY, not the sink. Add validation or rejection at the hop where untrusted data should have been checked — not a band-aid at the sink.
- Re-scan. Confirm the finding is gone.
Exit codes (CLI path)
0— clean (or gate not requested).1— the gate tripped: a non-suppressed defect at/above--fail-on.2— a wardline error (bad config, unreadable path). Not a finding.
Branch on the code. On a trip, read the structured report wardline just wrote — the finding names the function, file, and lines, which is enough to locate the leak.
Suppression discipline
Prefer FIXING a finding. Suppress only a finding you have judged a true non-issue, always with a reason:
- MCP
baseline— snapshot current defects so only NEW findings surface.overwrite: false(default) refuses to clobber an existing baseline;overwrite: truere-derives it. A coarse, whole-set tool; requires a reason. waiver_add— waive ONE finding by fingerprint with a mandatory reason and an expiry date. An audited, time-boxed exception.wardline judge(opt-in, network) — an LLM pass that labels each defect TRUE/FALSE positive. Never runs automatically, never folded into scan; fails loud with no API key so "couldn't triage" is never mistaken for "nothing to triage". Above-floor false positives can be recorded as audited suppressions.
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.
- 1mo ago First seen · 69 lines · 69 tokens per session scan A 52505ea4ecdf
wardline-gate is a skill published in the GitHub repository foundryside-dev/plainweave (0 stars, last pushed 3mo ago), licensed MIT. It adds 69 tokens to every session and 881 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-31.
Other skills, from other repositories
aomi-transact
Build natural-language crypto/DeFi agents and EVM MCP plugins (Claude Code, Cursor, Codex, Gemini). Aomi turns prompts into wallet-signed txs on Ethereum, Base, Arbitrum, Optimism, Polygon, Linea — non-custodial, fork-simulated. 40+ apps: Uniswap, Aave, Lido, Morpho, GMX, Hyperliquid, Polymarket.
grill-me
Stress-test a plan, design, or decision when the user explicitly asks to be grilled, challenged, or interviewed about assumptions, trade-offs, and unresolved requirements.
derive-nfr-from-driver
Derives measurable NFRs from a parent business driver.
derive-security-from-risk
Derives security requirements from a named threat.
requirements
Requirements-engineering expert that operates in three modes: (1) Elicitation — extract atomic, testable requirements from briefs, meeting notes, or system specs using EARS notation, with full attribute set (ID, type, priority, ASIL, verification method, source), flagging ambiguities as open questions; (2) Refinement…
plan
Structured technical planning before implementation — architecture and task breakdown.