skills/appmap-research-librarian/SKILL.md
Use when creating a gated research campaign for AppMap where one agent scouts external sources, another validates them into structured technique packs, and AppMap later ingests only reviewed local seed data or explicit validated URLs.
npx skillsauth add ghostonbutterbread/bug-bounty-harness appmap-research-librarianInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Create an offline campaign workspace that bridges open-ended source research and deterministic AppMap ingest.
/appmap-research-librarian init <program> --category <class> [--research-query WORD [WORD ...]] [--target-kind <kind>] [--focus rce]
/appmap-research-librarian validate <campaign_dir> [--seed <validated_seed.json>]
/appmap-research-librarian plan-appmap <campaign_dir> <target_path> [--write-specs] [--output-mode standalone|canonical] [--family <family>] [--lane <lane>] [--use-web-sources]
Examples:
/appmap-research-librarian init canva --category electron-ipc --research-query electron rce --target-kind electron-exe
/appmap-research-librarian validate ~/Shared/appmap/canva/research-librarian/<run_id>
/appmap-research-librarian plan-appmap ~/Shared/appmap/canva/research-librarian/<run_id> /home/ryushe/Shared/binaries/canva/exe/input/app_asar --write-specs --output-mode canonical --family binaries --lane exe
Read the playbook first:
$HARNESS_ROOT/prompts/appmap-research-librarian-playbook.md$HARNESS_ROOT/prompts/appmap-playbook.md when planning AppMap ingest$HARNESS_ROOT/agents/appmap_research_librarian.py$HARNESS_ROOT/prompts/appmap-research-librarian-playbook.md~/Shared/appmap/{program}/research-librarian/{run_id}/{campaign}/scout_brief.md{campaign}/validator_brief.md{campaign}/sources.todo.jsonl{campaign}/validated_research_seed.json{campaign}/validation_report.json{campaign}/plan_appmap_command.txtinit with the program and category/class.scout_brief.md.sources.todo.jsonl.validator_brief.md.validated_research_seed.json with cited sources and technique_packs.validate; fix any validation_report.json errors. Planning requires at least one validated source and one validated technique pack.plan-appmap; prefer local seed mode unless the user explicitly wants validated URL web mode.--research-mode local --research-seed <validated_seed> for replayable campaigns.--use-web-sources plans --research-mode web only from validator-approved HTTPS source URLs.testing
Systematic live request mutation: flip booleans, field ops, headers, content-type, parser differentials, replay vs intercept, null/empty testing. Inherits live-testing-policy scope/rate/ownership rules.
development
Test password reset, forgot-password, reset-token, email reset, and account recovery flows for account takeover risks.
tools
Targeted param/field discovery using tech stack clues, naming conventions, and controlled-rate ffuf — then feeds findings into request-exploration for mutation. Not brute-force; informed and scoped.
testing
Ghost-only workflow for creating approved bug bounty test accounts and saving credential references.