Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ericrisco/rsc-harness --skill linkedin-contentgit clone --depth 1 https://github.com/ericrisco/rsc-harnessWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ericrisco/rsc-harness/linkedin-content)<a href="https://agentmods.dev/skills/ericrisco/rsc-harness/linkedin-content"><img src="https://agentmods.dev/badge/skills/ericrisco/rsc-harness/linkedin-content.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00138 | $0.03822 |
| Opus 5 | $0.00069 | $0.01911 |
| Sonnet 5 | $0.00028 | $0.00764 |
| Haiku 4.5 | $0.00014 | $0.00382 |
Grade A, and why
linkedin-content scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LinkedIn Content — The Words a Human Pastes Into the Feed
You write the copy of one LinkedIn post. You take a raw idea, a story, or an asset and turn it into finished text the user pastes straight into the composer. You do not pick the topic or the day, you do not design the carousel pixels, you do not write DMs, and you do not call the API. You hand over words.
The one rule that governs every line: the 2026 algorithm pays for dwell time and comments, not likes or clicks. That dwell drives ranking is not a marketing claim — LinkedIn's own engineering team documents it: they train a "Long Dwell" classifier and feed per-post dwell into the ranking model precisely because dwell captures the passive readers that likes miss (LinkedIn Engineering, "Leveraging Dwell Time to Improve Member Experiences on the LinkedIn Feed", Oct 2024 — [S1]). The size of the gap is reported by practitioner analyses, not LinkedIn: posts with 0–3s dwell are reported to average ~1.2% engagement vs. ~15.6% at 61+s, a ~13x gap ([S2], corroborated by [S3]). Treat the mechanism as solid and the multiplier as directional. A 30-second read beats 50 quick likes. So every line you write has exactly one job: earn the next line, or earn the comment. If a sentence does neither, cut it. Why: dwell is the currency, and a line that doesn't pull the eye downward stops the clock.
Sources — what backs the numbers (and how hard)
Every figure below is keyed inline as [S#]. One primary source ([S1]) underpins the mechanism; the multipliers come from practitioner analyses and one named industry report, treated as directional, not gospel. All accessed 2026-06-02.
- [S1] — primary, authoritative. LinkedIn Engineering, "Leveraging Dwell Time to Improve Member Experiences on the LinkedIn Feed" (Oct 2024):
https://www.linkedin.com/blog/engineering/feed/leveraging-dwell-time-to-improve-member-experiences-on-the-linkedin-feed. LinkedIn's own write-up of the Long-Dwell classifier and dwell-aware ranking. Backs that dwell ranks and documents outdwell text — NOT the exact 13x/15x multipliers. - [S2] — practitioner analysis. dataslayer.ai, "LinkedIn Algorithm 2026: What Works Now (Documents, Newsletters, Video)":
https://www.dataslayer.ai/blog/linkedin-algorithm-february-2026-whats-working-now. Backs the ~60% body-link reach hit, the first-comment-penalty claim, the ~2–5% golden-hour test sample, the ~5% recovery rate, and the sub-60s video figure. - [S3] — practitioner analysis, second source. meet-lea.com, "LinkedIn Algorithm Explained 2026: Dwell Time, Comments & Reach":
https://meet-lea.com/en/blog/linkedin-algorithm-explained. Independently states the 1.2% vs. 15.6% dwell figures and the ~15x comment weight — and flags the ~15x as an industry estimate with AuthoredUp's quality-aware ~2x as the conservative alternative. - [S4] — named industry research report. Richard van der Blom, "LinkedIn Algorithm Insights Report 2026" (large-scale study, ~400k profiles):
https://richardvanderblom.com/. Corroborates the dwell-over-likes weighting, the in-body-link reach loss (~18.8% median for one link), and the link-in-first-comment suppression as of early 2026.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 163 lines · 138 tokens per session scan A d05b6f09c4e1
linkedin-content is a skill published in the GitHub repository ericrisco/rsc-harness (70 stars, last pushed today), licensed MIT. It adds 138 tokens to every session and 3,822 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
todos
This chat has a shared, live TODO plan — your tasks for the conversation, which the user also edits. Read this skill and reach for the todo tools whenever a request takes more than a couple of steps. It covers the plan model (group = task, items = its steps; loose items are the user's lane), how to work it: propose…
writing-workflow-skills
Use when adding a new workflow skill to pi-thinkrail-workflow, changing an existing workflow skill's role, trigger, handoff, or structure, or checking a workflow skill against the workflow system's rules. Not for authoring general-purpose skills outside this package.
setting-up-a-project
Use whenever asked to set up, onboard, initialize, or spec a project — the front door when the workspace has no spec graph yet (brand-new or an existing codebase); also seeded by the app's Set-up-project card (/skill:setting-up-a-project). Not for feature work in an already-specced project — use the brainstorming…
writing-specs
Use when a workflow step drafts or revises a spec artifact — a goal-and-requirements, an architecture, or a module SPEC — or when a workflow skill names it at such a step. The shared quality bar for specs — not a workflow, nothing to execute.
detecting-process-injection-techniques
Detects and analyzes process injection techniques used by malware including classic DLL injection, process hollowing, APC injection, thread hijacking, and reflective loading. Uses memory forensics, API monitoring, and behavioral analysis to identify injection artifacts. Activates for requests involving process…
extracting-iocs-from-malware-samples
Extracts indicators of compromise (IOCs) from malware samples including file hashes, network indicators (IPs, domains, URLs), host artifacts (file paths, registry keys, mutexes), and behavioral patterns for threat intelligence sharing and detection rule creation. Activates for requests involving IOC extraction, threat…