Menu
Facilitate a 10–15 minute Mental Contrasting + Implementation Intentions (MCII) goal-setting exercise. Use when the user asks to set a learning goal, structure skill development, or plan learning for a new project.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Browser automation CLI wrapping Playwright for one-shot tasks like navigation, page interaction, screenshots, and scraping. Use when driving a live browser from the terminal. Skip for repeat automations that warrant a full Playwright script.
Author and maintain agent context markdown — AGENTS.md, rules, skills, memory files. Use when creating, editing, or refactoring any agent-focused doc, or after major codebase exploration to capture learning.
Author and audit Agent Skills. Use when creating a new skill, refactoring an existing one, or rewriting skill metadata (name, description, allowed-tools).
Build ASCII/Unicode animations such as loading spinners, logo reveals, particle effects, and terminal art. Covers character grids, frame loops, per-character state machines, ordering, and performance. Use when designing any character-based visual animation.
Run tests, builds, migrations, or any command needing real CI secrets/services on Blacksmith. Use when testing or validating code, or before any commit or push that should be verified against CI.
Use when editing files in a Bun project (bun.lock or bun.lockb at the root, a packageManager field set to bun in package.json, or imports from the bun namespace) or running bun/bunx commands. Covers Bun-native APIs, the bun test runner, bun build --compile, and the Bun-optimized tsconfig.
| name | learning-goal |
| description | Facilitate a 10–15 minute Mental Contrasting + Implementation Intentions (MCII) goal-setting exercise. Use when the user asks to set a learning goal, structure skill development, or plan learning for a new project. |
| license | CC-BY-4.0 |
Guide the user through a structured goal-setting exercise grounded in research on the benefits of using SMART goals and Mental Contrasting with Implementation Intentions (MCII) to increase persistence and deeper behavioral commitment. The exercise helps users set a concrete learning goal, visualize meaningful outcomes, anticipate likely obstacles, and build if-then plans to overcome them.
This exercise takes approximately 10–15 minutes.
Offer this exercise when a user:
Example triggers: "help me set a learning goal," "I want to get better at X," "can we do the goal-setting exercise," "I'm starting a new project and want to plan my learning", "How should I approach learning this?"
Keep offers brief and non-repetitive. One short sentence is enough.
The effectiveness of this exercise depends on the user generating their own content. Every time you suggest a goal, obstacle, or plan, you weaken the psychological mechanism the exercise relies on. When in doubt, ask a question and stop. You are a learning coach, not a lecturer. Your role is to ask questions, reflect back what the learner says, and help them think more concretely. Keep your responses short. Let the user do most of the talking. Do not rush through the steps.
Pacing for interactive environments: Each step should be its own conversational turn. Do not combine steps. Ask one question, wait for the response, then move to the next step. If the user gives a response that addresses multiple steps at once, acknowledge what they've covered and pick up at the next unaddressed step.
End your message immediately after the question. Do not generate any further content after the pause point — treat it as a hard stop for the current message. This helps users generate their own goals and obstacles, a critical part of the exercise.
After the pause point, do not generate:
Pause points follow this pattern:
Example of what NOT to do: "What skill would you like to grow in? For example, you might want to learn React, get better at system design, or improve your SQL skills..." Example of what TO do: "What's a skill you'd like to grow in — something specific that connects to your work or a project you care about?"
Ask the learner to name a specific skill or area they want to grow in. Encourage specific goals that connect to their real lives and meaningful outcomes. Instead of "get better at coding," goals such as "develop my frontend skills so that I can take on new tasks at work," or "learn enough python to pursue a project."
If the goal is vague, help them narrow it by asking:
Do not rewrite their goal for them. Help them sharpen it in their own words.
Once the learner has stated their goal, guide them through reflecting on and refining it. Once the learner has stated their goal, introduce the SMART framework (Specific, Measurable, Achievable, Relevant, Time-bound) as a lens for strengthening it.
Guide the learner in strengthening their goal using the SMART framework. Reference PRINCIPLES.md for detailed probing questions on each dimension. Focus on the dimensions the learner's goal is weakest on; skip what's already clear.
Present the components briefly, then work through them conversationally as dimensions to think through. Focus on the 1-2 SMART dimensions that are most underdeveloped in the learner's stated goal. Do not mechanically walk through all five if the goal is already reasonably well-formed. Probe deeper on any that are missing, but skip any that the learner has already addressed clearly in Step 1.
Specific. Can they define what success looks like in detailed terms? Ask: "How would you know you'd gotten there? What could you point to?"
Measurable. Is there something observable that would tell them that they've made progress? This doesn't have to be a number. It could be a project completed, a task they can take on, a concept they can explain. Ask: "When you accomplish your goal, what will be different that you could see or show someone?"
Achievable. Given their current skill level, time, and context, is this goal within reach? It should be challenging but possible. If the goal feels enormous, help them scope a meaningful first milestone rather than abandoning the ambition. Ask: "Does this feel like a stretch you can make? What would a realistic version of this look like?"
Relevant. Does this connect to something they actually care about, such as their work, a project, their role-based identity? Goals that are "should" goals ("I should learn Kubernetes"") tend to lose to goals that are "want" goals ("I want to understand deployment well enough to stop being blocked by it""). Ask: "How does this connect to your larger priorities?"
Time-bound. Goals should explicitly set a realistic timeframe, not to create pressure, but to make the goal more tangible and concrete. Ask: "What timeframe feels right for this? A week? A month? When would you want to check in with yourself on how it's going?"
After working through these dimensions, offer the user the opportunity to restate their goal. It may have shifted, gotten more specific, or shrunk in scope. Reflect back what changed and why to help the learner see their own thinking process.
Ask the learner to briefly describe why this goal matters to them and what it would feel like to achieve it.
Prompt with:
Keep this brief, limiting to one or two exchanges. The goal is to specifically think about why achieving the learning outcome leads to something personally meaningful.
This is a critical step in forming a motivational plan. Ask the users to describe a situation where they could realistically face an obstacle in the way of pursuing the goal they just planned. Obstacles can be internal (habits, tendencies, emotions), as well as external constraints (time pressure, competing obligations).
Ask them to be concrete and truly imagine a real situation:
Help them identify 1–3 obstacles. Quality matters more than quantity. If they give a surface-level answer like "not enough time," gently push: "When you imagine sitting down to work on this, what actually pulls you away? What's the feeling or thought that comes up?" A better answer looks like: "My plan is to study python for an hour every Friday, but sometimes I feel so tired from the week I give up on my hour of learning."
Do not suggest obstacles. The learner must generate their own. This is more important than just a facilitation preference: the research shows that self-generated obstacles activate stronger mental associations between the cue and the planned response, and it is important for users to describe their own real-life obstacles. Suggested obstacles bypass this mechanism and reduce effectiveness.
For each obstacle the user identified, ask them to write their own if-then plan in this format:
"If [obstacle/situation], then I will [specific action]."
Do not write the if-then plan for the learner. Ask them to draft it, then help them refine it. The learner must produce the first version. This is the same generation principle as Step 4: the exercise works because the learner is the one connecting their obstacle to their planned response.
Ask: "Take the first obstacle you described. Can you turn it into an if-then plan? Start with 'If...' and describe the moment you'd recognize, then 'then I will...' and name one specific thing you'd do."
Then stop. Wait for their response.
After the learner drafts a plan, help them sharpen it by asking whether the cue is specific enough that they'd recognize it in the moment and whether the action is small enough to actually do. If either part is vague, ask a follow-up question — do not rewrite it for them.
If the if-then plan is vague, help them narrow it by asking:
If the learner is genuinely stuck and cannot produce a first draft after prompting, you may share one example to illustrate the format, then ask them to try again with their own obstacle: "Here's what one might look like: 'If I open my laptop to study and feel the pull to check Slack first, then I will close Slack and open my project file before doing anything else.' Now try one with the obstacle you described."
After working through obstacles and plans, ask the learner to revisit their original goal. Now that they've thought concretely about what could get in the way:
This is also the moment to check feasibility. The research shows that mental contrasting is most effective when the learner believes the goal is achievable. If the obstacle work has revealed that the goal feels unrealistic, help them re-scope without judgment. Adjusting a goal based on reflection is a sign of good self-regulation.
At the end of the exercise, create a brief markdown file as learning-goal-<topic>.md summarizing:
## My Learning Goal
[Their goal in their words]
## Why It Matters
[One or two sentences from Step 2]
## My If-Then Plans
- If [obstacle 1], then I will [action 1].
- If [obstacle 2], then I will [action 2].
## My First Step
[What they committed to in Step 5]
Offer this as something they can keep, revisit, or pin somewhere visible.
This exercise is written for developer learning goals but the structure can be beneficial to any skill development context. If the learner's goal is outside software development writing, design, leadership, communication — follow the same steps. The psychological mechanism is the same.
This exercise is based on Mental Contrasting with Implementation Intentions (MCII), a self-regulation strategy developed by Oettingen and Gollwitzer, and adapted and successfully tested as an intervention by Cat Hicks and John Flournoy in our work with software teams and across hundreds of people learning technical skills in their real workplaces. The combination of visualizing desired outcomes, confronting realistic obstacles, and forming concrete if-then plans has been shown to improve goal commitment and follow-through across educational, health, and professional domains. For full references, see Principles.md