Verify Agent Output
Validate deliverables and build evidence trails when work passes between agents. Covers expected outcome specification before execution, structured evidence generation during execution, deliverable validation against external anchors after execution, fidelity checks for compressed or summarized outputs, trust boundary classification, and structured disagreement reporting on verification failure. Use when coordinating multi-agent workflows, reviewing cross-agent handoffs, producing external-facing outputs, or auditing whether an agent's summary faithfully represents its source material.
Install
npx skills add pjt222/agent-almanac --skill verify-agent-outputgit clone --depth 1 https://github.com/pjt222/agent-almanac.git /tmp/agent-almanac
mkdir -p ~/.claude/skills
cp -r /tmp/agent-almanac/i18n/caveman-lite/skills/verify-agent-output ~/.claude/skills/verify-agent-outputPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Verify Agent Output Establish verifiable delivery between agents. When one agent produces output that another agent consumes — or that a human relies on — the handoff needs more than "looks good." This skill codifies the practice of defining checkable expectations before work begins, generating evidence as a side effect of doing the work, and validating deliverables against external anchors rather than self-assessment. The core principle: fidelity cannot be measured internally. An agent cannot reliably verify its own compressed output; verification requires an external reference point. ## When to Use - A multi-agent workflow hands deliverables from one agent to another - An agent produces external-facing output (reports, code, deployments) that a human will rely on - An agent summarizes, compresses, or transforms data and the summary must faithfully represent the source - A team coordination pattern requires structured handoff validation between members - You need to establish trust boundaries — deciding what requires verification vs. what can be trusted - An audit trail is required for compliance or reproducibility ## Inputs - **Required**: The deliverable to verify (file, artifact, report, or structured output) - **Required**: The expected outcome specification (what "done" looks like) - **Optional**: The source material (for fidelity checks on summaries or transformations) - **Optional**: Trust boundary classification (`cross-agent`, `external-facing`, `internal`)
Details
- Skill name
verify-agent-output- Repository
- pjt222/agent-almanac
- Path
i18n/caveman-lite/skills/verify-agent-output/SKILL.md- License
- MIT
- Repo updated
- 18 November 2026
- Listed in
- BehiSecc/awesome-claude-skills
FAQ
What does the Verify Agent Output skill do?
Validate deliverables and build evidence trails when work passes between agents. Covers expected outcome specification before execution, structured evidence generation during execution, deliverable validation against external anchors after execution, fidelity checks for compressed or summarized outputs, trust boundary classification, and structured disagreement reporting on verification failure. Use when coordinating multi-agent workflows, reviewing cross-agent handoffs, producing external-facing outputs, or auditing whether an agent's summary faithfully represents its source material.
How do I install the Verify Agent Output skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add pjt222/agent-almanac --skill verify-agent-output. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Verify Agent Output popular?
It lives in pjt222/agent-almanac, which has 45 GitHub stars and 4 forks, and is included in 1 curated skill list.
Related skills
Adapt Architecture
45pjt222/agent-almanac
Execute structural metamorphosis using strangler fig migration, chrysalis phases, and interface preservation. Covers transformation planning, parallel running, progressive cutover, rollback design, and post-metamorphosis stabilization for system architecture evolution. Use when assess-form has classified the system as READY for transformation, when migrating from monolith to microservices, when replacing a core subsystem while dependents continue operating, or when any architectural change must be gradual rather than big-bang.
Adaptic
45pjt222/agent-almanac
Master skill composing the 5-step synoptic cycle for panoramic synthesis across multiple domains. Orchestrates meditate, expand-awareness, observe, awareness, integrate-gestalt, and express-insight into a coherent process that produces unified understanding rather than sequential compromise. Use when a problem genuinely spans 3+ domains and the interactions between domains matter more than depth in any one, when sequential analysis feels like compromise rather than integration, or before major architectural decisions affecting multiple stakeholders.
Add Puzzle Type
45pjt222/agent-almanac
Scaffold a new puzzle type across all 10+ pipeline integration points in jigsawR. Creates the core puzzle module, wires it into the unified pipeline (generation, positioning, rendering, adjacency), adds ggpuzzle geom/stat layers, updates DESCRIPTION and config.yml, extends the Shiny app, and creates a comprehensive test suite. Use when adding a completely new puzzle type to the package or following the 10-point integration checklist to ensure nothing is missed end-to-end.
Version Ml Data
45pjt222/agent-almanac
Version machine learning datasets using DVC (Data Version Control) with remote storage backends, build reproducible data pipelines with dependency tracking, integrate with Git workflows, and ensure data lineage for model reproducibility. Use when versioning large datasets that do not fit in Git, tracking data changes alongside code changes, ensuring ML experiment reproducibility, sharing datasets across team members, or auditing data lineage for compliance requirements.
Vendor AI Review
43AlsKozlov/ru-legal
Review AI vendor contracts (LLM API providers, ML libraries, AI services). Включает SaaS-MSA общие issues (см. contract-law/saas-msa-review) + AI-specific: training data rights, output ownership, model warranty, hallucination liability, rate limits, API stability, безопасность (запрет training на our data), exit strategy + portability. Post-2022 sanctions specifics.
Write Claude Md
45pjt222/agent-almanac
Create an effective CLAUDE.md file that provides project-specific instructions to AI coding assistants. Covers structure, common sections, do/don't patterns, and integration with MCP servers and agent definitions. Use when starting a new project where AI assistants will be used, improving AI behavior on an existing project, documenting project conventions and constraints, or integrating MCP servers or agent definitions into a project workflow.
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.