Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Finding Synthesis · athola | DecimalAI

researchby athola

Finding Synthesis

athola/synthesize

Source athola/claude-night-market ↗License MIT

Merges, dedupes, ranks, and formats research findings into a report. Use after research agents return results from multiple channels to produce a ranked report.

SkillScore96Maintained

Automatic

Research pack

Last verified 8/21/2026

What it changes

Passed 22 of 23 test cases with the skill, and 3 of 23 without it.

Without skill13%

With skill96%

Checks it passed that the model alone missed

Explicitly states that 0 findings were merged.
Names tome.synthesis.ranker as the module path.
Calls format_brief to generate the brief output.

and 21 more in the full report

20 fixed2 passed either way0 still failing1 broke

Pass-rate lift: +83 pp
Tokens: -41%
Agent turns: 0%
Response time: -60%

Model gemini-3.6-flashMeasured 8/21/2026v122 of 23 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/athola-synthesize/SKILL.md https://app.decimal.ai/s/athola-synthesize/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add athola-synthesize, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/athola-synthesize/ and pins it in .decimal/skills.lock

decimalai skills add athola-synthesize

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/athola-synthesize/SKILL.md https://app.decimal.ai/s/athola-synthesize/SKILL.md

Raw file: /s/athola-synthesize/SKILL.md ↗ · machine-readable summary at /s/athola-synthesize.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 20 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 20

Prompt

What is the full Python function path used to format the final full markdown report in tome?

What was checked, without and with the skill
Checked	Without skill	With skill
Names tome.output.report.format_report as the report formatting path.	Missed	Passed

case-21 · gemini-3.6-flash · 8/21/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md39 linessingle file

Finding Synthesis

When To Use

After research agents return results from multiple channels
Producing a final ranked report from raw findings

When NOT To Use

No research session is active (run /tome:research first)
Refining a single channel (use /tome:dig instead)

Merge findings from all channels into a ranked report.

Workflow

Merge: tome.synthesis.merger.merge_findings()
Rank: tome.synthesis.ranker.rank_findings()
Group: tome.synthesis.ranker.group_by_theme()
Format: tome.output.report.format_report()

Output Formats

report: Full sectioned markdown
brief: Condensed 1-2 pages
transcript: Raw session log

Exit Criteria

merge_findings(), rank_findings(), and group_by_theme() all called in sequence before output is formatted
Output formatted via format_report() by default, or format_brief() / format_transcript() when the flag is set
Findings grouped by theme in the report, not presented as raw per-channel lists
If the merged finding count is 0, this is stated explicitly rather than generating an empty or fabricated report
If no active session exists, error emitted directing the user to run /tome:research first

Track record

Benchmark

22 / 23

test cases passed, vs 3 / 23 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

athola / claude-night-market

plugins/tome/skills/synthesize/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Finding Synthesis

When To Use

When NOT To Use

Workflow

Output Formats

Exit Criteria

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills