- Home
- Skills
- Marketing & SEO
- Signal Mining: Content Angles Hidden in Raw Material
More screenshots
Works with the AI Agents you already use
Signal Mining: Content Angles Hidden in Raw Material
Triage a messy dump of news, threads, and competitor posts into the few content angles your brand genuinely has standing to publish, with every rejection written down.
$5
Signal Mining: Content Angles Hidden in Raw Material
Example session with this skill installed
I run a fictional bike shop brand called Copperfield Cycles, selling to city commuters. Pillars: city commuting, bike maintenance, family cycling. Audience in one line: city commuters who ride daily and fix their own bikes. Here is a fictional dump: a rival shop announced a big sale, the city proposed a new bike-lane network, a survey claims 58% of commuters skip winter riding, and a thread of riders complains winter routes are never cleared. Mine it and show me the angles, the dropped list, and what needs verifying. Invent plausible details.
- Read your context and instructions
- Compiled the signal mining
SIGNAL MINE — Copperfield Cycles, invented details for this demo only.
SPLIT: rival sale (event); city bike-lane proposal (event plus its proposed routes); 58% skip winter riding (survey number); winter routes never cleared (rider complaint thread).
SCORING: Rival sale — no pillar home, competitor's story, dropped. Bike-lane proposal — pillar city commuting; standing strong, since Copperfield Cycles staff ride the proposed routes daily and can name which ones actually connect; window this month while the proposal is open for comment; facts checkable from the proposal text. Kept. 58% figure — pillar city commuting; standing medium, since shop repair logs show a winter drop-off, but the number itself is pasted with no publication attached. Verify first, not a draft. Winter routes thread — pillar city commuting; standing strong, since the shop's winter maintenance clinics already teach riding on uncleared routes; window this month; framed as observed complaints, not a statistic. Kept.
ANGLES: A1. "The proposed bike-lane network fixes the wrong gap first — here are the three routes commuters actually need." Pillar city commuting, format article, window this month. Why this brand: staff ride these routes daily and can compare the proposal against real commute patterns. A2. "Uncleared winter routes are a maintenance problem, not a reason to stop riding." Pillar city commuting, format post, window this month. Why this brand: its winter clinics already teach exactly this riding.
DROPPED: rival sale — competitor's story, no standing. FIRST-RUN FAULTS, FIXED: the first pass kept the rival sale as an angle ("what big sales mean for local shops") and treated the 58% figure as a fact. The authority-fit question dropped the sale — commenting hands the rival the frame — and the claim-risk question moved the 58% figure to the verify list until its own publication is checked.
VERIFY BEFORE DRAFTING: the 58% figure, against the survey's own publication, before it appears in any draft.
Connects securely to your tools. The creator never sees your data.
About this skill
Most content ideas fail not because they are boring but because they are not yours: off-pillar trend-chasing, commentary on a competitor's story, or a take any brand in the space could have written. This skill mines a raw dump the disciplined way. It starts with three prerequisites — the dump itself, the brand's three to five pillars, and the audience in one line — and stops cold if the pillars are missing. The dump is then split into one-sentence signals, each with a source tag, and every signal is scored in writing against four questions: pillar fit, authority fit, timeliness against real production speed, and claim risk. Pillar fit is necessary and authority fit is decisive; a signal that passes the first and fails the second is the most common rejection, and the skill says so plainly. Survivors are shaped into angles — a take in two sentences, pillar and format, a window matched to real ship speed, and one sentence of standing — while everything else lands on the dropped list with its reason, so any rejection can be overruled deliberately. Verification stays a separate step: pasted numbers never become cited facts until checked from their own source. The output template and a pre-delivery gate close the loop, including the honest outcome of a mine with zero survivors.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Marketing & SEO.
- It cannot mine without defined pillars; without them it stops rather than guessing.
- It does not verify claims itself; verification is a separate step done by someone with access to the source.
- It assumes the dump is pasted material; it does not fetch live news or social feeds.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 days ago
Permissions
No special permissions declared or detected