research-agent/skills/synthesize/SKILL.md
Force a research-agent run to conclude. Launches the researcher in synthesis mode: it reads state.md and log.md, writes .research/synthesis.md with outcome, key findings, failed approaches, open questions, and recommendations, then exits. Use when current results are good enough or the agent is stalling.
npx skillsauth add queelius/claude-anvil synthesizeInstall 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.
Force the researcher to write its final synthesis and conclude, even if it would otherwise continue exploring.
Use the Read tool on .research/state.md. If the file does not exist, tell the user there is no research to synthesize and suggest /research-agent:research to start one.
Use the Glob tool for .research/synthesis.md. If it already exists, ask the user whether they want to overwrite (re-synthesize, perhaps because more cycles have happened since) or stop.
Spawn the researcher agent (Agent tool, subagent_type research-agent:researcher) with this prompt:
<mode>synthesize</mode>
<working_directory>
{absolute path to current working directory}
</working_directory>
Synthesize your research and conclude. Do NOT run new attempts. Do
NOT start new cycles. The user has decided to lock in current
results.
Step 1: Read .research/state.md and .research/log.md in full to
recall every sub-problem, hypothesis, and attempt.
Step 2: Update .research/state.md with final statuses. Mark every
sub-problem as one of: resolved, abandoned, or unresolved. Mark
every hypothesis as one of: confirmed, refuted, inconclusive.
Step 3: Write .research/synthesis.md per the Termination Protocol
in your system prompt. Required sections: Goal, Outcome (one
paragraph), Key Findings, Failed Approaches, Open Questions,
Recommendations.
Step 4: Append a final entry to .research/log.md noting the forced
conclusion and pointing at synthesis.md.
Be honest in the synthesis. If the goal was not achieved, say so
plainly. If results are partial, describe their scope precisely.
The synthesis is the single document a future reader will rely on.
After launching, tell the user:
Researcher synthesizing. It will read state.md and log.md, finalize statuses, and write .research/synthesis.md. This usually completes in one cycle (no new attempts).
data-ai
Show the current state of an in-flight research-agent run from .research/state.md, log.md, and attempts/. Read-only summary of cycles, sub-problems, hypothesis statuses, eval trend, and current focus.
testing
Resume an interrupted research-agent run. Re-launches the researcher with instructions to read .research/state.md and log.md, reorient, and continue from the documented current focus. Use after a context compression, session restart, or explicit pause.
tools
When and how to use the repoindex plugin surface (MCP tools, agents, slash commands) for collection queries, release prep, activity summaries, and tag discipline. Use when users ask repoindex questions, mention their repo catalog, or want to know which repoindex tool fits their task.
development
Use when working in a project with a worldbuilding documentation ecosystem (docs/ or lore/ directory). Provides the documentation-first methodology: canonical hierarchy, doc roles (timeline, lore, characters, systems), propagation discipline, and the dual canonical and exploratory workflow.