Monitors competitor websites, pricing, content changes, hiring patterns, and product updates. Generates intelligence reports with strategic implications and trend analysis. Stores history for longitudinal tracking.
SkillScore80Maintained
Automatic
◆Product Management pack
Last verified 8/24/2026
What it changes
Passed 20 of 25 test cases with the skill, and 6 of 25 without it.
Without skill24%
With skill80%
Checks it passed that the model alone missed
Updates `usage-history.json` with the run metadata.
Triggers an immediate alert for changes scored 4 or 5.
Delivers actionable intelligence with historical context.
and 17 more in the full report
15 fixed5 passed either way4 still failing1 broke
Pass-rate lift
+56 pp
Tokens
+26%
Agent turns
0%
Response time
-7%
Model gemini-3.6-flashMeasured 8/24/2026v120 of 25 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 15 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 15
Prompt
After finishing a monitoring run across 3 competitors, what specific log file must be updated with the execution metadata and run timestamp?
What was checked, without and with the skill
Checked
Without skill
With skill
Updates `usage-history.json` with metadata from the monitoring run.
Missed
Passed
case-12 · gemini-3.6-flash · 8/24/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md32 lines5 files
Competitor Intelligence Agent
Track competitor activity across multiple dimensions, detect meaningful changes, interpret the signals, and deliver actionable intelligence that builds historical context over time. Act as an analyst that connects dots, not a raw scraper.
Contents
references/directory-structure.md -- tracking directory layout, config.yaml, and usage-history.json templates
references/monitoring-dimensions.md -- the six monitoring dimensions with per-dimension analysis frameworks, detection protocols, and snapshot output formats
references/intel-report-format.md -- the full intelligence report template
Setup (no tracking directory exists): collect the user's company name and description, competitor URLs/domains, priority monitoring dimensions, and output directory (default ./competitor-intel/). Create the directory structure and config.yaml. See references/directory-structure.md.
Monitoring run (tracking directory exists): proceed to steps 2-7.
Report only (user wants a report without new monitoring): read existing snapshots and change logs, synthesize trends, and generate strategic recommendations using references/intel-report-format.md.
Read config.yaml to load the competitor list and settings, then read the most recent snapshot for each competitor and dimension.
Execute monitoring across all configured dimensions. Apply the detection protocol for each dimension in references/monitoring-dimensions.md.
Compare new data against previous snapshots. Score every change for magnitude per references/scoring-and-rules.md; flag changes rated 4-5 as immediate alerts.
Write dated snapshots in the per-dimension output formats and log detected changes under the competitor's changes/ folder.
Generate the intelligence report following references/intel-report-format.md. When 3 or more snapshots exist for a competitor, add longitudinal trend analysis.
Update usage-history.json with the run metadata.
Guardrails
Never fabricate competitor data. If a fetch fails or a dimension has no data, state the gap.
Separate raw data (snapshots) from interpretation (reports).
Tag every data point with source, timestamp, and confidence; flag data older than 30 days as stale.
Recommend only legal, ethical competitive responses. Collect only publicly available professional information.
Apply the detailed change-detection, trend, data-quality, and execution rules in references/scoring-and-rules.md throughout.