Use when a founder is the bottleneck, can't delegate, is burning out or is moving from IC to CEO. Covers delegation, energy and calendar audits, blind spots and succession.

Use before a board meeting, investor update or fundraising pitch where the numbers will be challenged. Drills numbers cold, anticipates the hard questions and builds a narrative that admits weakness without losing the room.

Check sales pipeline and forecasts: pipeline coverage, forecast accuracy, magic number, CAC payback. Use whenever the user asks if pipeline is enough or how good forecasts are.

Use when a launched managed agent needs grading against a rubric and improving ('grade my agent', 'is it good enough'). Builds the outcome with outcome_builder.py, reads verdicts with verdict_reader.py and runs held-back evals from eval_scaffold.py.

Use when someone wants to build, launch, grade or schedule a Claude Managed Agent in their own Anthropic account ('build me an agent', 'run this nightly'). Reads the session goal, routes with goal_router.py and compiles the loop with loop_compiler.py.

Use for GDPR or DSGVO work: scanning a codebase for privacy risks, writing a DPIA or tracking data subject requests against the Art. 12(3) deadline. Runs gdpr_compliance_checker.py, dpia_generator.py and data_subject_rights_tracker.py; final calls go to the DPO.

Learn or apply deep reinforcement learning with OpenAI Spinning Up: policy gradients, VPG, PPO, SAC, TD3, DDPG, TRPO. Use whenever an RL agent will not learn, reward stays flat, or the user asks how PPO or SAC works.

Use for growing on X/Twitter: tweets, threads, profile audit, competitors or posting plan. Audits the profile with profile_auditor.py, drafts with tweet_composer.py and plans posts with content_planner.py.

Plan A/B tests and product experiments: hypotheses, sample size, priority, reading results. Use whenever the user asks if a test won, how long to run it or what to test.

Write a blog post, article or guide end to end: research, outline, draft, SEO pass and quality gates. Use whenever the user asks to write a blog post, draft an article on a topic, or turn notes into a full piece.

Use for LinkedIn comments, connection notes, DMs or outreach plans, or checking outreach volume is safe. Plans targets with comment_target_planner.py, drafts notes with outreach_message_builder.py and checks limits with outreach_volume_guard.py.

Use when setting up email triage for the first time, before inbox-triage, or when priorities change. Interviews the user one question at a time and writes the knowledge base inbox-triage reads, validated with kb_validator.py.

Blunt market-first verdict on an idea, venture or career bet, leading with the strongest counterargument, or a 3x5-card daily plan. Use whenever the user asks is there a market, are we at PMF, should I do this, or says pmarca mode.

Use when an autonomous agent takes a side-effecting action (deploy, delete, pay, grant access) that may need proving later, or for EU AI Act Art. 12 record-keeping. Builds a tamper-evident signed receipt with build_action_manifest.py and verifies it.

Use when the user shares meeting transcripts (.vtt, .srt, .txt, Otter, Granola, Zoom) or asks how they come across in meetings. Measures speaking ratio, filler words and avoidance patterns and gives coaching.

Use when exposing an API as an MCP server or scaffolding a new MCP server in Python or TypeScript, before hand-writing tool wrappers. Generates tools from OpenAPI with openapi_to_mcp.py and validates them with mcp_validator.py.

Turn a markdown file into a single-file HTML slide deck with keyboard navigation, presenter notes and print-to-PDF. Use whenever the user wants slides or a presentation from markdown or notes.

File-by-file pull request review for bugs, security and quality, with a merge verdict. Load whenever the user asks to review, look over or check a PR or diff before merging.

Use when preparing or running a user access review, checking that leavers lost their access, finding orphaned or ownerless service accounts, answering an auditor's access-control request for SOC 2, ISO 27001, PCI DSS, HIPAA or NIST 800-53, or checking whether a finished review campaign will hold up. Diffs the HR roster against account exports, tests the campaign for coverage, self-review, unremoved revokes and rubber-stamping, and maps every exception to the controls it breaks.

Use before installing any skill or plugin from an untrusted source, or when asked if a skill is safe. Scans scripts, SKILL.md and dependencies for exfiltration, dangerous calls and prompt injection with skill_security_auditor.py.