Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →larscowe/bmad-distillator
Lossless LLM-optimized compression of source documents.
Passed 7 of 15 test cases with the skill, and 2 of 15 without it.
and 6 more in the full report
8 fixed2 passed either way11 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 7 test cases the skill fixed. Same model, same prompt, with and without the skill.
Please condense the policy guides located at `./compliance/2024/` so they can be consumed by a downstream regulatory audit bot. I need a lossless distillation that preserves all key facts and section headings. Also, run a full round-trip reconstruction test against the original files to verify no critical data was lost during compression. Return the final structured JSON object detailing the status, generated file paths, token usage metrics, and completeness results.
| Checked | Without skill | With skill |
|---|---|---|
| The round-trip validation executes using the --validate workflow path. | Missed | Passed |
| A validation report file named with the suffix -validation-report.md is generated containing information preservation metrics and gap assessments. | Missed | Passed |
The instructions your model reads when the skill loads.
This skill produces hyper-compressed, token-efficient documents (distillates) from any set of source documents. A distillate preserves every fact, decision, constraint, and relationship from the sources while stripping all overhead that humans need and LLMs don't. Act as an information extraction and compression specialist. The output is a single dense document (or semantically-split set) that a downstream LLM workflow can consume as sole context input without information loss.
This is a compression task, not a summarization task. Summaries are lossy. Distillates are lossless compression optimized for LLM consumption.
Validate inputs. The caller must provide:
-distillate.md suffix.Route — proceed to Stage 1.
| # | Stage | Purpose |
|---|---|---|
| 1 | Analyze | Run analysis script, determine routing and splitting |
| 2 | Compress | Spawn compressor agent(s) to produce the distillate |
| 3 | Verify & Output | Completeness check, format check, save output |
| 4 | Round-Trip Validate | (--validate only) Reconstruct and diff against originals |
Run scripts/analyze_sources.py --help then run it with the source paths. Use its routing recommendation and grouping output to drive Stage 2. Do NOT read the source documents yourself.
Single mode (routing = "single", ≤3 files, ≤15K estimated tokens):
Spawn one subagent using agents/distillate-compressor.md with all source file paths.
Fan-out mode (routing = "fan-out"):
test cases passed, vs 2 / 15 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.