Content Creator — Commitment System Setup
You are configuring the commitments system for a content creator. Their day involves:
- Morning: scanning trends and planning
- Midday: creating (writing, filming, recording)
- Afternoon: editing, thumbnails, publishing
- Evening: distribution across platforms, audience engagement
- Ideas arrive constantly and most won't be executed immediately
Persistence rules
For creator workflows, persistence comes before presentation.
- When the user asks to track a new content piece, update pipeline progress,
add a deadline, capture a trend reaction, or log a sponsored obligation, you
must write the corresponding workspace file before confirming success.
- Do not say a content item, trend response, distribution task, or sponsored
deadline was "tracked", "created", "queued", or "added to the pipeline"
unless the
memory_write succeeded.
- If the user asks only to track deadlines or commitments, do not generate
assets, thumbnails, scripts, tweets, or video copy unless they explicitly
ask for that creative work.
- Trend reactions like "create a short take on React compiler tonight" should
normally become active commitments in
projects/commitments/open/, not just prose or
a suggestion.
Concrete examples:
- "Track this commitment: create a short take on the React compiler trend tonight."
Required behavior: write an active commitment under
projects/commitments/open/ before confirming.
- "Track this: the sponsored Figma workflow video has to ship by Friday."
Required behavior: write a sponsored commitment under
projects/commitments/open/.
- "Track this commitment only: I need TikTok cuts and a Twitter thread for Episode 48 by tomorrow morning."
Required behavior: write the deadline commitment(s); do not generate scripts, threads, or assets unless explicitly asked.
Companion skills
This bundle relies on these skills activating during conversation (keyword-triggered):
| Skill |
Activates when |
What it does |
commitment-triage |
User mentions obligations, deadlines |
Extracts signals, creates/resolves commitments |
commitment-digest |
User asks "show commitments" |
Composes formatted summary |
decision-capture |
User makes a decision |
Records decision with rationale |
idea-parking |
User says "park this idea" |
Parks ideas for weekly resurfacing |
If any are missing from skills/, tell the user which ones are needed.
Step 1: Ask configuration questions
- Platforms: Which platforms do you publish to? (YouTube, TikTok, Instagram, Twitter, blog, podcast, etc.)
- Content cadence: How often do you publish? (daily, 2-3x/week, weekly)
- Sponsored content: Do you have sponsored/partner content with hard deadlines?
- Trend sensitivity: How quickly do trends expire for your niche? (hours, days)
Step 2: Declare the project and create workspace structure
Writing any file under projects/commitments/ is the declaration that
the project exists — the engine auto-registers it and scopes missions
to it. Start with:
memory_write(
target: "projects/commitments/AGENTS.md",
content: "# Commitments (Creator)\n\nThis project tracks content pipeline items, publishing deadlines, and parked ideas for a content creator.\n\n## Operating principles\n\n- Pipeline stages: idea → research → script → create → edit → thumbnail → publish → distribute → engage.\n- Trend-related signals expire after 6h — trends move fast.\n- Sponsored content due within 3 days is always `urgency=critical`.\n- When the user publishes on one platform, auto-create distribution commitments for the others (only after explicit setup).\n",
append: false
)
Then:
- Check if
projects/commitments/README.md exists via memory_read. If it does, skip to creating the content-pipeline directory.
- Write
projects/commitments/README.md with the full schema — see commitment-setup skill for the complete content including immediacy levels, signal destinations, resolution paths, and trust calibration.
- Create placeholder READMEs in each subdirectory:
open/, resolved/, signals/pending/, signals/expired/, decisions/, parked-ideas/.
- Create the content-pipeline directory:
memory_write(target="projects/commitments/content-pipeline/README.md", content="# Content Pipeline\n\nEach content piece gets its own file tracking its lifecycle:\nidea → research → script → create → edit → thumbnail → publish → distribute → engage\n\nFiles: projects/commitments/content-pipeline/<slug>.md\n\nWhen a piece is published on one platform, create distribution commitments for the other platforms automatically.", append=false)
Step 3: Create tuned missions
Triage mission — trend-aware, fast expiration
mission_create(
name: "commitment-triage",
goal: "Creator triage. Read projects/commitments/README.md for schema. Priority order: (1) Sponsored content with hard deadlines — flag anything due within 3 days as urgency=critical. (2) Content pipeline items in projects/commitments/content-pipeline/ — check for stalled stages (not updated in 2+ days). (3) Trend-related signals (obligation_type with 'trend' in tags) — expire after 6 hours if not promoted (trends move fast). Non-trend signals expire after 48 hours. (4) For signals with immediacy=realtime (viral moment, platform outage), broadcast immediately via message. (5) Check parked-ideas/ for ideas that might be timely now based on recent signals. (6) Route informational signals (industry news, competitor moves) to intelligence via MemoryDoc. (7) Append triage summary to projects/commitments/triage-log.md. (8) Alert if any sponsored deadlines are approaching.",
cadence: "0 8,14,20 * * *",
project_id: "commitments"
)
Digest mission — pipeline-focused
mission_create(
name: "commitment-digest",
goal: "Creator digest. Read projects/commitments/README.md for schema. Sections: (1) CONTENT IN PROGRESS — list items from projects/commitments/content-pipeline/ with their current stage and days since last update. Flag stalled items. (2) SPONSORED DEADLINES — any commitments tagged 'sponsored' with due dates. (3) PUBLISHING QUEUE — items in 'publish' or 'distribute' stage. For agent_can_handle items (scheduling posts, writing descriptions), offer to proceed. (4) FRESH IDEAS — count of parked ideas, highlight high-relevance ones from the last week. (5) ENGAGEMENT TASKS — commitments about responding to comments, collaborations. End with 'Did I miss anything?' Send via message tool.",
cadence: "0 8 * * *",
project_id: "commitments"
)
Idea resurface mission — weekly
mission_create(
name: "creator-idea-resurface",
goal: "Weekly parked ideas review for content creator. Read all files in projects/commitments/parked-ideas/ via memory_tree and memory_read. For ideas parked more than 2 weeks ago, compose a brief list asking if they are still interesting. For high-relevance ideas, suggest promoting them to the content pipeline. If any parked ideas align with recent trending signals, highlight the match. Send the list via message tool. If no parked ideas exist, skip silently.",
cadence: "0 10 * * 1",
project_id: "commitments"
)
Step 4: Write calibration memories
memory_write(
target: "projects/commitments/calibration.md",
content: "# Content Creator Calibration\n\n- Content pieces are tracked as pipeline items in projects/commitments/content-pipeline/, not as plain commitments\n- Pipeline stages: idea → research → script → create → edit → thumbnail → publish → distribute → engage\n- When user publishes on one platform, automatically create commitments for distribution to other platforms: <platforms list>\n- Trend-related signals expire after 6 hours — if not acted on quickly, they are stale\n- Sponsored content is always urgency=critical when due within 3 days\n- Ideas flow constantly — park liberally, promote selectively\n- Parked ideas are resurfaced weekly on Monday mornings\n- When a new content piece starts, create a pipeline file with all stages as unchecked items\n- For agent_can_handle items (scheduling, descriptions, thumbnails), ask permission before proceeding\n- Start conservative: surface everything, learn preferences over time",
append: false
)
Replace <platforms list> with the platforms the user listed in Step 1.
Step 5: Confirm
Your content creator system is ready:
- Triage runs 3x daily (8am, 2pm, 8pm) — trend signals expire in 6h, sponsored deadlines flagged at 3 days
- Morning digest at 8am — pipeline status, deadlines, publishing queue, fresh ideas
- Idea resurface every Monday morning — reviews parked ideas older than 2 weeks
- Pipeline tracking in
projects/commitments/content-pipeline/ — each piece tracks idea through engagement
- Cross-platform cascades: tell me when you publish and I'll create distribution commitments
- Say "new content piece: [title]" to start a pipeline, or "park this idea" to save for later
Step 6: Mark setup complete
After confirming with the user, write the setup completion marker so this skill stops competing for the activation budget on every subsequent message:
memory_write(
target: "projects/commitments/.content-creator-setup-complete",
content: "# Content Creator Setup Complete\n\nCompleted: <today's UTC date>\n\nMissions installed: creator-triage, creator-digest, creator-idea-resurface"
)
To re-trigger setup, delete projects/commitments/.content-creator-setup-complete first.
1---
2name: content-creator-setup
3version: 0.3.0
4description: One-time onboarding for the content creator workflow — content pipeline stages, trend expiration, cross-platform cascades, heavy idea parking. After successful setup this skill is excluded from selection until the marker file is deleted.
5activation:
6 setup_marker: projects/commitments/.content-creator-setup-complete
7 keywords:
8 - content creator
9 - creator assistant
10 - youtube workflow
11 - content pipeline
12 - publishing schedule
13 - creator setup
14 - content calendar
15 patterns:
16 - "(?i)I'm a (content creator|youtuber|creator|streamer|podcaster|blogger)"
17 - "(?i)set ?up.*(content|creator|publishing|video)"
18 - "(?i)help me manage my (content|videos|publications|posts)"
19 tags:
20 - commitments
21 - content-creation
22 - publishing
23 - setup
24 max_context_tokens: 2500
25requires:
26 skills:
27 - commitment-triage
28 - commitment-digest
29 - decision-capture
30 - idea-parking
31metadata:
32 source: https://skillmd.com/skills/gabrielmoreira/content-creator-setup
33---
34
35# Content Creator — Commitment System Setup
36
37You are configuring the commitments system for a content creator. Their day involves:
38- Morning: scanning trends and planning
39- Midday: creating (writing, filming, recording)
40- Afternoon: editing, thumbnails, publishing
41- Evening: distribution across platforms, audience engagement
42- Ideas arrive constantly and most won't be executed immediately
43
44## Persistence rules
45
46For creator workflows, persistence comes before presentation.
47
48- When the user asks to track a new content piece, update pipeline progress,
49 add a deadline, capture a trend reaction, or log a sponsored obligation, you
50 must write the corresponding workspace file before confirming success.
51- Do not say a content item, trend response, distribution task, or sponsored
52 deadline was "tracked", "created", "queued", or "added to the pipeline"
53 unless the `memory_write` succeeded.
54- If the user asks only to track deadlines or commitments, do not generate
55 assets, thumbnails, scripts, tweets, or video copy unless they explicitly
56 ask for that creative work.
57- Trend reactions like "create a short take on React compiler tonight" should
58 normally become active commitments in `projects/commitments/open/`, not just prose or
59 a suggestion.
60
61Concrete examples:
62- "Track this commitment: create a short take on the React compiler trend tonight."
63 Required behavior: write an active commitment under `projects/commitments/open/` before confirming.
64- "Track this: the sponsored Figma workflow video has to ship by Friday."
65 Required behavior: write a sponsored commitment under `projects/commitments/open/`.
66- "Track this commitment only: I need TikTok cuts and a Twitter thread for Episode 48 by tomorrow morning."
67 Required behavior: write the deadline commitment(s); do not generate scripts, threads, or assets unless explicitly asked.
68
69## Companion skills
70
71This bundle relies on these skills activating during conversation (keyword-triggered):
72
73| Skill | Activates when | What it does |
74|---|---|---|
75| `commitment-triage` | User mentions obligations, deadlines | Extracts signals, creates/resolves commitments |
76| `commitment-digest` | User asks "show commitments" | Composes formatted summary |
77| `decision-capture` | User makes a decision | Records decision with rationale |
78| `idea-parking` | User says "park this idea" | Parks ideas for weekly resurfacing |
79
80If any are missing from `skills/`, tell the user which ones are needed.
81
82## Step 1: Ask configuration questions
83
841. **Platforms**: Which platforms do you publish to? (YouTube, TikTok, Instagram, Twitter, blog, podcast, etc.)
852. **Content cadence**: How often do you publish? (daily, 2-3x/week, weekly)
863. **Sponsored content**: Do you have sponsored/partner content with hard deadlines?
874. **Trend sensitivity**: How quickly do trends expire for your niche? (hours, days)
88
89## Step 2: Declare the project and create workspace structure
90
91Writing any file under `projects/commitments/` is the declaration that
92the project exists — the engine auto-registers it and scopes missions
93to it. Start with:
94
95```
96memory_write(
97 target: "projects/commitments/AGENTS.md",
98 content: "# Commitments (Creator)\n\nThis project tracks content pipeline items, publishing deadlines, and parked ideas for a content creator.\n\n## Operating principles\n\n- Pipeline stages: idea → research → script → create → edit → thumbnail → publish → distribute → engage.\n- Trend-related signals expire after 6h — trends move fast.\n- Sponsored content due within 3 days is always `urgency=critical`.\n- When the user publishes on one platform, auto-create distribution commitments for the others (only after explicit setup).\n",
99 append: false
100)
101```
102
103Then:
104
1051. Check if `projects/commitments/README.md` exists via `memory_read`. If it does, skip to creating the content-pipeline directory.
1062. Write `projects/commitments/README.md` with the full schema — see `commitment-setup` skill for the complete content including immediacy levels, signal destinations, resolution paths, and trust calibration.
1073. Create placeholder READMEs in each subdirectory: `open/`, `resolved/`, `signals/pending/`, `signals/expired/`, `decisions/`, `parked-ideas/`.
1084. Create the content-pipeline directory:
109
110```
111memory_write(target="projects/commitments/content-pipeline/README.md", content="# Content Pipeline\n\nEach content piece gets its own file tracking its lifecycle:\nidea → research → script → create → edit → thumbnail → publish → distribute → engage\n\nFiles: projects/commitments/content-pipeline/<slug>.md\n\nWhen a piece is published on one platform, create distribution commitments for the other platforms automatically.", append=false)
112```
113
114## Step 3: Create tuned missions
115
116### Triage mission — trend-aware, fast expiration
117
118```
119mission_create(
120 name: "commitment-triage",
121 goal: "Creator triage. Read projects/commitments/README.md for schema. Priority order: (1) Sponsored content with hard deadlines — flag anything due within 3 days as urgency=critical. (2) Content pipeline items in projects/commitments/content-pipeline/ — check for stalled stages (not updated in 2+ days). (3) Trend-related signals (obligation_type with 'trend' in tags) — expire after 6 hours if not promoted (trends move fast). Non-trend signals expire after 48 hours. (4) For signals with immediacy=realtime (viral moment, platform outage), broadcast immediately via message. (5) Check parked-ideas/ for ideas that might be timely now based on recent signals. (6) Route informational signals (industry news, competitor moves) to intelligence via MemoryDoc. (7) Append triage summary to projects/commitments/triage-log.md. (8) Alert if any sponsored deadlines are approaching.",
122 cadence: "0 8,14,20 * * *",
123 project_id: "commitments"
124)
125```
126
127### Digest mission — pipeline-focused
128
129```
130mission_create(
131 name: "commitment-digest",
132 goal: "Creator digest. Read projects/commitments/README.md for schema. Sections: (1) CONTENT IN PROGRESS — list items from projects/commitments/content-pipeline/ with their current stage and days since last update. Flag stalled items. (2) SPONSORED DEADLINES — any commitments tagged 'sponsored' with due dates. (3) PUBLISHING QUEUE — items in 'publish' or 'distribute' stage. For agent_can_handle items (scheduling posts, writing descriptions), offer to proceed. (4) FRESH IDEAS — count of parked ideas, highlight high-relevance ones from the last week. (5) ENGAGEMENT TASKS — commitments about responding to comments, collaborations. End with 'Did I miss anything?' Send via message tool.",
133 cadence: "0 8 * * *",
134 project_id: "commitments"
135)
136```
137
138### Idea resurface mission — weekly
139
140```
141mission_create(
142 name: "creator-idea-resurface",
143 goal: "Weekly parked ideas review for content creator. Read all files in projects/commitments/parked-ideas/ via memory_tree and memory_read. For ideas parked more than 2 weeks ago, compose a brief list asking if they are still interesting. For high-relevance ideas, suggest promoting them to the content pipeline. If any parked ideas align with recent trending signals, highlight the match. Send the list via message tool. If no parked ideas exist, skip silently.",
144 cadence: "0 10 * * 1",
145 project_id: "commitments"
146)
147```
148
149## Step 4: Write calibration memories
150
151```
152memory_write(
153 target: "projects/commitments/calibration.md",
154 content: "# Content Creator Calibration\n\n- Content pieces are tracked as pipeline items in projects/commitments/content-pipeline/, not as plain commitments\n- Pipeline stages: idea → research → script → create → edit → thumbnail → publish → distribute → engage\n- When user publishes on one platform, automatically create commitments for distribution to other platforms: <platforms list>\n- Trend-related signals expire after 6 hours — if not acted on quickly, they are stale\n- Sponsored content is always urgency=critical when due within 3 days\n- Ideas flow constantly — park liberally, promote selectively\n- Parked ideas are resurfaced weekly on Monday mornings\n- When a new content piece starts, create a pipeline file with all stages as unchecked items\n- For agent_can_handle items (scheduling, descriptions, thumbnails), ask permission before proceeding\n- Start conservative: surface everything, learn preferences over time",
155 append: false
156)
157```
158
159Replace `<platforms list>` with the platforms the user listed in Step 1.
160
161## Step 5: Confirm
162
163> Your content creator system is ready:
164> - **Triage** runs 3x daily (8am, 2pm, 8pm) — trend signals expire in 6h, sponsored deadlines flagged at 3 days
165> - **Morning digest** at 8am — pipeline status, deadlines, publishing queue, fresh ideas
166> - **Idea resurface** every Monday morning — reviews parked ideas older than 2 weeks
167> - Pipeline tracking in `projects/commitments/content-pipeline/` — each piece tracks idea through engagement
168> - Cross-platform cascades: tell me when you publish and I'll create distribution commitments
169> - Say **"new content piece: [title]"** to start a pipeline, or **"park this idea"** to save for later
170
171## Step 6: Mark setup complete
172
173After confirming with the user, write the setup completion marker so this skill stops competing for the activation budget on every subsequent message:
174
175```
176memory_write(
177 target: "projects/commitments/.content-creator-setup-complete",
178 content: "# Content Creator Setup Complete\n\nCompleted: <today's UTC date>\n\nMissions installed: creator-triage, creator-digest, creator-idea-resurface"
179)
180```
181
182To re-trigger setup, delete `projects/commitments/.content-creator-setup-complete` first.