research ยท diff

git:20260715.f00df98 to git:20260723.6e5370b

76 added, 33 removed. Audit A to A.

---
name: research
description: Produce bounded, source-backed research packets for product, ecosystem, and operator decisions.
runx:
category: research
---
# Research
- Research one bounded question and turn it into a decision-ready packet.
+ Answer one practical question with enough evidence to change a decision. This
+ skill is for applied research: which issue is worth tackling, whether a proposal
+ is grounded, what changed in an ecosystem, or which claims a public artifact can
+ safely make. It is not an open-ended browsing session or a summary for its own
+ sake.
- This skill is for applied research, not open-ended browsing. It should answer
- one practical question with evidence, tradeoffs, and explicit uncertainty:
- which issue is worth tackling, what the ecosystem is doing, whether a proposal
- is grounded, or what claims a public post can safely make.
+ The result should make the operator's posture clearer: build, write, avoid,
+ defer, compare, or gather a named missing fact. Prefer a small number of
+ well-supported findings over a broad report whose conclusions cannot be traced.
- Keep the scope tight. Summaries without evidence are not enough, but an
- undirected literature review is also wrong. Prefer a small number of verified
- claims that change the operator's decision.
+ ## Where it fits
- ## Operating rules
+ `research` consumes governed source packets. Fetch or read evidence first with
+ `web-fetch`, a provider-specific reader, or another skill that returns stable
+ source and content digests. URLs, model memory, operator context, and target
+ names do not become evidence merely because they were supplied.
- - State the objective in operational terms.
- - Distinguish verified evidence from inference.
- - Give every important claim a source and confidence.
- - Surface missing evidence instead of inventing it.
- - Bound the result to a concrete deliverable: brief, issue recommendation,
- content outline, or publish/no-publish decision.
- - State what the finding changes: what to write, build, avoid, defer, or review.
- - Return `needs_more_evidence` rather than forcing a speculative conclusion,
- and `not_worth_publishing` when a true finding is irrelevant to the audience.
+ The packet is a reusable primitive for `deep-research`, `content-pipeline`,
+ `ecosystem-brief`, `ghostwrite`, product planning, and other decision workflows.
+ It does not publish, notify, or mutate a provider. Those later operations keep
+ their own authority, approval, and readback boundary.
+ ## How it works
- ## Output
+ 1. State a bounded objective and the decision the research should inform.
+ 2. Supply up to the declared source limit as governed readback packets. Runx
+ indexes their provenance, observation times, source digests, and provider
+ content digests before synthesis.
+ 3. Synthesis separates verified facts from inference, explains relevance,
+ records confidence, and carries open questions rather than filling gaps with
+ plausible prose.
+ 4. Every material claim cites an admitted source digest. Every recommended
+ option cites the evidence on which its rationale depends.
+ 5. The deterministic verifier rejects unknown citations, digest drift, and
+ unsupported decision support before the packet is released.
- - `research_brief`: object with `objective`, `scope`, `summary`, and
- `open_questions`.
- - `evidence_log`: array of evidence entries with `claim`, `source`,
- `confidence`, and `relevance`.
- - `decision_support`: array of options or recommendations with rationale.
- - `risks`: array of research or execution risks.
+ Return `needs_more_evidence` when the admitted sources cannot support the
+ decision. Return `not_worth_publishing` when a finding is sound but irrelevant
+ to the stated deliverable. Neither state is a failure; both prevent weak
+ evidence from turning into confident downstream content.
- ## Inputs
+ ## Inputs and result
- - `objective` (required): the question to answer.
- - `domain` (optional): ecosystem, product area, or audience context.
- - `deliverable` (optional): intended artifact, for example `daily brief`,
- `triage recommendation`, or `publish packet`.
- - `operator_context` (optional): local constraints or strategic context.
- - `target_entities` (optional): array or object naming repos, products,
- competitors, communities, or issues that bound the research.
+ - `objective` is the exact question to answer.
+ - `source_packets` are bounded provider or fetch readback packets with stable
+ provenance and content digests.
+ - `domain`, `deliverable`, `operator_context`, and `target_entities` narrow the
+ analysis. They are interpretation context, never source evidence.
+
+ The research packet contains the bounded brief, evidence log, decision options,
+ risks, open questions, admitted source evidence, and deterministic validation.
+ Consumers should carry the packet and its digest forward rather than rewriting
+ its findings into an untraceable summary.
+
+ ## Stop conditions
+
+ - Stop when no valid source packet was supplied, evidence is stale for the
+ decision, or source identity cannot be verified.
+ - Never cite an unknown source digest or treat a URL as proof that it was read.
+ - Label inference as inference and lower confidence when evidence conflicts.
+ - Do not manufacture balance by inventing an opposing claim with no source.
+ - Keep source content out of public artifacts unless the downstream skill is
+ allowed to use it and respects quotation and privacy constraints.
+ - Route requests to browse, publish, message, or mutate to the skill that owns
+ that boundary.
+
+ ## Example
+
+ An operator asks whether a proposed integration belongs in the next release and
+ supplies a changelog readback, an issue snapshot, and an official API document.
+ The packet can compare the evidence, state which user problem is verified, and
+ recommend build, defer, or investigate. It cannot cite a remembered forum post
+ or operator hunch as verified demand, and it should name the missing evidence if
+ that gap changes the recommendation.
+
+ ## Agent task contract
+
+ ### `research-synthesize`
+
+ Synthesize only the supplied source index. Preserve the objective exactly and
+ return a bounded brief, evidence log, decision support, and risks. Each material
+ claim must cite admitted source and provider-content digests; each recommendation
+ must cite the source digests behind its rationale. Distinguish verified evidence
+ from inference and return `needs_more_evidence` when support is insufficient.
+ Do not browse, publish, or invent provider, delivery, or certification claims.