agentsclimarketplace

Piensalo write

Skill ralfyishere/piensalo/skills/piensalo-write

Writing and content-creation program: build the claim ledger and constraint checklist first, draft divergently, run separate argument/accuracy/style critic passes, compress without losing ledger items. Use for high-stakes prose - announcements, notices, abstracts, memos, docs - where every claim must be defensible and omissions are costly. Trigger phrases: 'draft', 'write the announcement/memo/notice', 'tighten this'.From its SKILL.md

Install
npx -y skills add ralfyishere/piensalo --skill piensalo-write

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

4.8 KB, 937 tokens by cl100k_base, as published. Nobody here has run it

piensalo-write

A structured reasoning program for writing and content creation, distilled from curated expert reasoning traces. Work the numbered steps in order; each step's output feeds the next. Do not skip steps silently - if one does not apply, say so in one line.

Primary workflow

  1. Recover real objective - Per intent-clarity: decode what the requester actually needs before optimizing anything. Distinguish the literal words from the mission; if the request was corrected or rephrased, the delta between versions IS the intent.
  2. Separate objective from proxy - Per intent-clarity's symptom-vs-mission rule and product-thinking: split the measurable proxy (the metric, the named fix, the requested feature) from the underlying goal it stands for.
  3. Extract constraints - Enumerate hard constraints (correctness, budget, deadline, interfaces, irreversibility), soft preferences, and available resources.
  4. Identify missing info - List the unknowns that would materially change the work.
  5. Audience objective model - Model the reader before writing a word: who they are, what they already know, what they will DO after reading, and under what attention budget.
  6. Constraint ledger - Compile the writing constraint ledger: length ceiling, register/voice, must-include facts, must-not-say items (confidentiality, legal, publication-hygiene boundaries), house style, format requirements.
  7. Structural alternatives - Per divergent-ideation applied to form: generate 3-5 genuinely different structures BEFORE drafting — inverted pyramid, narrative arc, problem/solution, FAQ, comparison table, letter — different skeletons, not one outline with renamed headings.
  8. Select structure - Choose the structure by one criterion above all: which gets the reader to reader_action fastest with the least attention spent (structured-reasoning: name the frame — this is a tradeoff between completeness and speed-to-action, and the reader's priority order beats the order you thought of things in).
  9. Draft divergence - Draft 2-3 versions with judgment OFF (divergent-ideation's divergence phase, applied to prose): vary the opening move, the voice temperature, and what gets compressed vs dwelt on.
  10. Critic passes - Switch from author to attacker (adversarial-verify, for prose). Run distinct passes, in order, per ruthless-editor: (1) STRUCTURE — lede first, one point per paragraph, kill the on-ramp ('In today's fast-moving landscape...'); (2) ARGUMENT — does the text establish what it claims, and does the opening promise only what the body delivers; (3) ACCURACY — check every factual claim against the ledger and recompute numbers independently; (4) STYLE — register, voice, passive constructions, hedge stacks.
  11. Compression pass - Per ruthless-editor's cut-and-strengthen passes: delete repetition, filler ('it's worth noting that', 'in order to'), hedge stacks (pick one hedge or none), and scaffolding sentences that only announce other sentences.
  12. Independent final verification - Verify the FINAL artifact, not the plan: execute every bound check method from determine_verifiables against the refined solution.
  13. Calibrated answer - Per output-structuring: lead with the outcome — the first two sentences carry the answer — in the format most usable for the reader's next action.

Conditional moves

  • When argument-pass failure: substance wrong, not style: return to Draft divergence and rework from there.
  • When compression altered meaning or dropped a ledger item: return to Critic passes and rework from there.
  • When final check found an unverified factual claim: return to Critic passes and rework from there.

Output contract

A finished piece that opens with its point, fits the constraint ledger exactly (length, voice, must-include, must-not-say), carries evidence-level labels on factual claims, plus the critique log and the structural alternatives considered.

Delivery notes (small-model packets)

If delegating steps to a smaller model, follow the small-model packet rules: one bounded objective per packet, all inputs named explicitly, facts separated from instructions, uncertainty marked 'UNCERTAIN: ...' rather than invented away.

Before answering

Check the draft against references/failure-checks.md (the failure-mode catalog) and references/verification.md (domain verification criteria). Repair procedures live in references/contracts.md.

What ships with it: 9 files

35.2 KB alongside SKILL.md

evals/

examples/

references/

Keep looking

Skills are one crate of 326,546. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.