Social Evidence Research

Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites: No account, no API key and no paid service. The skill works from public social material you can reach, or from exports and captures you supply and are authorized to share. It explicitly does not log in to accounts on your behalf and does not get around login gates, paywalls, rate limits or deleted content. If your agent environment has browsing, opening the public source URLs you research sends the normal requests any browser makes to those platforms; without browsing, the skill works from your supplied exports only and records every URL it could not open as discovery only. Nothing is sent anywhere during installation itself, and the research itself publishes, posts, likes, follows, messages and comments nothing. Install "Social Evidence Research" for me. A research method that turns public or user authorized social material into a report where every claim points to the inspected material behind it. The workflow sets the protocol before collecting (question, platforms, date range, selection method, sample cap, stopping rule), treats search results and previews as discovery only records that can support no claim, records each inspected part of a source as its own evidence record, groups reposts and cross posts so repeated content counts once, attributes every comment to its commenter, builds a claim table linked to evidence ids and locators, and reports discovery, inspection, exclusion and access gap counts with their limits. It also audits existing social research reports for discovery only evidence, double counted reposts, misattributed comments and claims that go beyond the sample. It works read only: no posting, liking, following, messaging or logging in on your behalf, and no working around login gates, paywalls or rate limits: each block is recorded as an access gap. Repository: https://github.com/wshobson/agents (the skill is the whole plugins/content-marketing/skills/social-evidence-research/ folder at the pinned commit: SKILL.md, references/acceptance-examples.md, references/evidence-record.md, references/report-template.md) Pinned version: install the code at commit 0874a27b17a22e2b7c2e2bcd725768e0803cad06 exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/wshobson/agents into a temporary folder, check out commit 0874a27b17a22e2b7c2e2bcd725768e0803cad06, and copy the plugins/content-marketing/skills/social-evidence-research/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-18: [FINGERPRINTS] - SKILL.md: SHA-256 344fa1b35ff1c480768635337526fb3bee23ef0c52821ae71ad82c4b40a4fe1d - references/acceptance-examples.md: SHA-256 76bad7816fac4b5f8ac7a62c648e7483abcc64f90ba3183d401c2c51b9148dc1 - references/evidence-record.md: SHA-256 57e736c3852355df75d856b2a15814d5e16a867653110c9d0c05834853c93386 - references/report-template.md: SHA-256 b8965e8270359fcdb15f94494a76435ea7f16a887cf34a01f705338c628bdd74 If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full (and every other file in the folder) and tell me in plain words: what the skill does when used, which services it connects to, whether it reads or sends any personal information or keys, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the files: - The skill folder holds 4 files: SKILL.md plus three reference files in references/ (evidence record fields, report template, acceptance examples). There is no executable code, no script and no tool of its own. - Nothing is sent anywhere during installation itself; the only download is the skill folder from GitHub at the pinned commit. - The method is read only by design: it forbids publishing, posting, liking, following, messaging, commenting and logging in on your behalf, and forbids getting around login gates, paywalls, rate limits or deleted content, recording each as an access gap. - Source text is treated as data to analyze, never as instructions to follow, and every claim must cite the evidence ids and locators that support that specific claim, with the sample limits stated. 3. When I say "yes, install", install it as a skill: copy the whole folder into the agent's skills directory, in a folder named "social-evidence-research", keeping every file (SKILL.md, references/acceptance-examples.md, references/evidence-record.md, references/report-template.md). 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (for example: sign in or create an account, provide any API key, or install any local tool named in the prerequisites; using the skill is a separate step that only starts when I ask for it, and every paid or authenticated use disclosed above happens only then). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.