Borrowing it
Nothing to install: this file belongs to re-cinq/wave. 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/re-cinq/wave/main/.agents/skills/wave-ctx-validation/SKILL.mdgit clone --depth 1 https://github.com/re-cinq/waveWrote 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/re-cinq/wave/wave-ctx-validation)<a href="https://agentmods.dev/skills/re-cinq/wave/wave-ctx-validation"><img src="https://agentmods.dev/badge/skills/re-cinq/wave/wave-ctx-validation/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/re-cinq/wave/wave-ctx-validation"><img src="https://agentmods.dev/badge/skills/re-cinq/wave/wave-ctx-validation.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 | $0.00090 | $0.01630 |
| Opus 5.5 | $0.00036 | $0.00652 |
| Sonnet 5.5 | $0.00018 | $0.00326 |
| Haiku 4.5 | $0.00009 | $0.00163 |
Grade A, and why
wave-ctx-validation 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Validation Context
Multi-layer output validation system — contract enforcement with six validator types (json_schema, typescript_interface, test_suite, markdown_spec, format, non_empty_file), progressive JSON recovery with three aggressiveness tiers, adaptive retry with failure classification and repair prompts, error wrapper detection for AI-generated envelopes, input artifact schema validation, preflight dependency checks with auto-install, and pipeline-level phase/concurrency/staleness validation.
Invariants
- Unknown contract types silently pass — NewValidator returns nil, Validate returns nil
- must_pass controls whether contract failure blocks the pipeline — true=hard error, false=soft failure event
- maxRetries<=0 defaults to 1 attempt (no retry) in basic mode, but 3 in adaptive mode — intentional asymmetry
- After exhausting all retries, Retryable is set to false on the ValidationError
- JSON schema validation requires either Schema (inline) or SchemaPath (file) — neither produces non-retryable error
- Test suite validator rejects unresolved template variables — '{{ ' or '{{' in command produces non-retryable error
- Test suite defaults Dir to 'project_root' resolved via git rev-parse --show-toplevel
- TypeScript validator degrades gracefully when tsc absent — must_pass:true = hard error with install instructions, must_pass:false = silently skips
- JSON recovery defaults to progressive when not explicitly configured and must_pass is false — conservative for must_pass contracts
- Recovered JSON must be structurally complete (balanced braces/brackets) — incomplete recovery is retryable error
- Error wrapper detection requires error_type AND raw_output AND at least 3 indicator fields
- Input artifact validation fails fast — first invalid artifact stops all further checks
- Non-empty file validator requires file exists AND has size > 0
- Missing tools return ToolError with list of missing tool names
- Missing skills return SkillError with list of missing skill names
- PreflightError wraps both ToolError and SkillError, supporting errors.As() for either
- Skills with Install command are auto-installed — post-install verification retries with $HOME/.local/bin in PATH
- Init commands NOT run in preflight — deferred to worktree creation
- Phase skip validation only applies to impl-prototype or prototype pipelines
- Generic step sequence validation requires prior run state to exist before resuming from a step
- on_failure:rework requires rework_step to be set — validated at parse time
- Workspace locks prevent concurrent access — in-memory lock returns error if workspace in use
- Optional steps default to on_failure:continue; non-optional default to on_failure:fail
- Format type inferred from schema filename — github-pr-draft.schema.json maps to github_pr
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 · 105 lines · 90 tokens per session scan A bc8ba27a861f
wave-ctx-validation is a skill published in the GitHub repository re-cinq/wave (23 stars, last pushed 1mo ago), licensed MIT. It adds 90 tokens to every session and 1,630 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
backend
Backend development skill for microservices following Clean Architecture, CQRS with a transactional outbox, the repository/use-case pattern, and a Testing-Trophy strategy.
tdd-workflow
Test-Driven Development workflow for implementing features with tests first; enforces the two-commit audit trail (failing-test commit, then green commit) and shows-your-work verification for agent-driven work.
verification
Structured multi-phase verification of implementations before deployment, covering static analysis, testing, functional checks, security, and documentation.
verification-before-completion
Comprehensive five-level validation checklist that every agent must pass before declaring any task or work package complete.
interactive-live-testing
Drive Playwright against a running application step-by-step, screenshot, analyze, decide, act. Applies to any web frontend regardless of stack.
Verification & Quality Assurance
Comprehensive truth scoring, code quality verification, and automatic rollback system with 0.95 accuracy threshold for ensuring high-quality agent outputs and codebase reliability.